Science topic

Autonomous Vehicles - Science topic

Explore the latest questions and answers in Autonomous Vehicles, and find Autonomous Vehicles experts.
Questions related to Autonomous Vehicles
  • asked a question related to Autonomous Vehicles
Question
5 answers
When will we have autonomous vehicles in use at Level 4: High Driving Automation or Level 5: Full Driving Automation?
What are the main problems? What are the road conditions? How will they adjust to different weather conditions? How will they share information with the road or with other vehicles? What technologies, algorithms, and processing detect and prevent the occurrence of accidents?
  • asked a question related to Autonomous Vehicles
Question
3 answers
Hello
I am a PhD student looking to read some recent good papers that can help me identify a research topic in RL for controls applications . I have been reading through quite a few papers/topics discussing model free vs model based RL etc . Not been able to find something , may be I don't understand it yet to the extent :) .
Just for the background : My experience is with Diesel , SI engines , vehicles and controls .
One of the topics/areas that seems interesting to me is learning using RL in uncertain scenarios, this might seem to broad for most of the people .
Another possible area would be RL for connected vehicles, self driving etc .
Any help/suggestion is welcome .
Relevant answer
Answer
combining MARL and safety would be an interesting area
  • asked a question related to Autonomous Vehicles
Question
3 answers
Hi there,
I am working on the development of sensors for Autonomous vehicles that can assess road conditions, particularly traffic situations. The primary area is Optics based sensors such as LiDAR or Photonic Radar. I am looking for tools to develop the sensor and asking for recommendation on which software is helpful.
Regards
Relevant answer
Answer
Optisystem from Optiwave Company
  • asked a question related to Autonomous Vehicles
Question
3 answers
Photonic-based or microwave photonic-based Sensors are being used in combination with Ultrasonic, and Camera-based sensors for the smooth operation of Autonomous Vehicles. Many articles have been presented using FMCW techniques along with LiDAR tech to enhance the sensing capabilities of AVs.
What do you think is the future direction in this area of sensor development in a simulative environment?
Relevant answer
Answer
Technology Integration via;
1. Joint Sensing and Communication,
2. Dual Function Radar Communication.
  • asked a question related to Autonomous Vehicles
Question
7 answers
We are planning to learn Apollo open source tool to simulate various scenario for autonomous vehicles. Could you please provide good learning resources for the same. Thank you.
Relevant answer
Answer
Can you explain more about the topic, please?
  • asked a question related to Autonomous Vehicles
Question
4 answers
Suppose I have and HD map, which have many segments (line, arc, spiral, …), that describe the road of the vehicle.
Now, How can I plot or get the coordinates along the spiral given the all following data:
  • start coordinate of the spiral,
  • the heading (orientation) of the spiral,
  • the start and end of the curvature for the spiral, and
  • the length of the spiral.
Relevant answer
Answer
Dear @Qamar Ul Islam, thanks for your response, however it is not the correct. Since the video explained the polar spiral or Archimedes spiral.
  • asked a question related to Autonomous Vehicles
Question
2 answers
Sensors have played a significant role in intelligent robots and autonomous driving. While lidars and cameras have become dominant in the field, the advancements of other sensors have also promoted its developments.
MMWR is becoming promising. Its 4D imaging technology is becoming more and more mature. Even though I am not familiar with this kind of sensor and have no experience with it, as a person working in robotic and autonomous systems, I am interested in it. I hope this message can reach someone who knows about it.
Questions are:
1. How about the role of 4D MMWR in robots and autonomous vehicles, especially compared with lidar and cameras? or even replace them as primary sensors?
2. Is 4D MMWR robust enough for L3 or above-level autonomous driving?
3. Are there any other fields that 4D MMWR will be applied more often or promising?
Relevant answer
Answer
1. How about the role of 4D MMWR in robots and autonomous vehicles, especially compared with lidar and cameras? or even replace them as primary sensors?
It can substitue lidas or camera depending on the application.
2. Is 4D MMWR robust enough for L3 or above-level autonomous driving?
For Level 3 and up I would always use redundant sensors, preferentially radar, camera and lidar.
3. Are there any other fields that 4D MMWR will be applied more often or promising?
Several applications. There is no general answer.
  • asked a question related to Autonomous Vehicles
Question
4 answers
Are there any available datasets of autonomous vehicles trajectories with GPS and date times specifications? if not, how would one adapt existing datasets, which are obtained from normal vehicles, to autonomous ones if possible?
Relevant answer
Answer
Here are same datesets for vehicle trajectory prediction/planning:
Argoverse
Appoloscape
NGSIM
HighD
nuScenes
CommonRoad
  • asked a question related to Autonomous Vehicles
Question
2 answers
A number of proprietary (and often expensive) softwares for modelling both vehicle dynamic behaviour, and the sensors for autonomous vehicles exist.
We have experience using open source solutions including Gazebo, AirSim, LG SVL (which sadly appears to no longer be supported by LG), CARLA, and a number of custom solutions bridging software like MATLAB to games engines like Unity.
Does anyone have any additional suggestions, or anyone experienced in this area have any insights / resources etc into what the future directions for coupled vehicle dynamic and autonomous vehicle simulation may be?
Relevant answer
Answer
For high-fidelity vehicle dynamics simulation, the gold standard should always be CarSim or ASM. For virtual environment creation, Unity may dominant.
  • asked a question related to Autonomous Vehicles
Question
14 answers
Hello,
I am a young researcher and preparing myself for master's related to image processing, computer vision and ai for autonomous vehicle.
I have worked on various projects but confused with their publication that which one will be better in conference paper, journal or the book chapter. It will be very supportive if anyone suggest the best one?
Relevant answer
Answer
The key difference between journal and conference paper is that a journal article is a long piece of writing that is published in journals and periodicals, whereas a conference paper is a concise and precise written paper that is presented in research conferences.
Both journals and conference papers present the findings and conclusion of a research study. Moreover, both journal articles and conference papers undergo a review process before the acceptance of papers.
Next, borrowing from Ian Kennedy: The conference paper gives faster feedback. The journal paper counts for about twice as much credit, but takes longer to get into press (an average of 9 months).
  • asked a question related to Autonomous Vehicles
Question
1 answer
I am studying at the minimum number of shared autonomous vehicles would need to meet all traffic demand of motor vehicles needs for a city, and considering the impact of joining rail transit to further optimize fleet size. I plan to cite the paper titled "Addressing the minimum fleet problem in on-demand urban mobility" in my paper, and compare the method of solving the minimum fleet size in this paper with my research method, so as to prove whether my research is effective. Therefore, I would like to ask if you have the source code for this article. If so, please contact me at Chengminghui@zjnu.edu.cn. Thank you very much!
Relevant answer
I recommend you contact the first author of the paper. You can find the email in the following page: https://www.nature.com/articles/s41586-018-0095-1
  • asked a question related to Autonomous Vehicles
Question
10 answers
Hello,
I have been working on computer vision. I used datasets from Kaggle or other sites for my projects. But now I want to do lane departure warning, and real-time lane detection with real-time conditions(illuminations, road conditions, traffic, etc.). Then the idea to use simulators comes to my mind but there are lots of simulators on online but I'm confused about which one would be suitable for my work!
It would be very supportive if anyone guide me through picking up the best simulator for my works.
  • asked a question related to Autonomous Vehicles
Question
9 answers
[Information] Special Issue - Intelligent Control and Robotics
Relevant answer
Answer
Thanks for sharing.
  • asked a question related to Autonomous Vehicles
  • asked a question related to Autonomous Vehicles
Question
9 answers
For most drivers who manually control a vehicle, obeying traffic regulations like speed limits looks like a huge task for them as some overlook the importance of these strategically placed limits. Our quest is to devise ways to use AI-based technology in both autonomous and manually controlled cars to detect these speed limits on the road and for the AI technology to proceed to act on it in bringing the vehicle to a lower speed when the vehicle is travelling at a speed out of range. Issuing warning about a speed out of range is another thing and it all falls on the driver to heed to the warnings or not.
Relevant answer
Answer
Hi Eric Ntiamoah . I agree with Qamar Ul Islam . Thanks.
  • asked a question related to Autonomous Vehicles
Question
7 answers
Considering the use of facial recognition cameras on autonomous vehicles, the storage of data collected, the security of the data and access, do UK data privacy laws go far enough to protect peoples rights?
Relevant answer
Answer
Great question - given the attitude afforded by the Investigatory Powers Act in the UK, there is a huge privacy concern posed by CAVs. These are essentially mobile surveillance systems and the concern is how footage from the onboard interior and exterior facing cameras is used and whether state bodies can request access to these. Lots of really questions surrounding CAV use and development in the UK.
Very best wishes, James.
  • asked a question related to Autonomous Vehicles
Question
4 answers
Hi friends! Who knows taxi drivers? (traditional cab drivers, not Uber of Lyft). Our team is recruiting taxi drivers to participate in focus groups about technological changes in driving jobs and the future impact of autonomous vehicles on workforce. Whoever successfully particpated in the focus group can receive a $50 gift card. More information can be found at: <iframe src="https://www.facebook.com/plugins/post.php?href=https%3A%2F%2Fwww.facebook.com%2FWEAVENSF%2Fposts%2F180149620718317&show_text=true&width=500" width="500" height="737" style="border:none;overflow:hidden" scrolling="no" frameborder="0" allowfullscreen="true" allow="autoplay; clipboard-write; encrypted-media; picture-in-picture; web-share"></iframe>
Relevant answer
Answer
Good to Know, Good Luck.
Kind Regards
Qamar Ul Islam
  • asked a question related to Autonomous Vehicles
Question
5 answers
Hello,
I am writing my thesis regarding building trust in autonomous vehicles (AVs). I have developed 4 hypothesis from literature review and my research model contains 4 different independent variables (Reliability, Usability, Familiarity, Regulations) and a single dependent variable (trust in AVs). I am using 5 point Likert scale based on previous studies to collect data through an online survey questionnaire. Each variable has 4 items.
With limited knowledge, I am not sure which tests would be necessary and enough to see which of my IVs really impact DV i.e. Trust?
Thank You
Shahid
Relevant answer
Answer
You can use SPSS or STATA or even statgraphics to code the data, analyze the data using multiple regression
  • asked a question related to Autonomous Vehicles
Question
3 answers
Since I am working on autonomous vehicle, I am interested in developing a path planning/navigation system for guiding in row crops. Please tell me about the type of algorithm/programme that needs to be developed for same. Also, inform me what type of camera and GPS system (Make; Model) for real time application will be the best that can identify the crop rows.
Is there any firm in India which can provide the necessary equipment/components.
Relevant answer
Answer
You are welcome. As I had already underlined, I have no knowledge about such cameras. I have never seen such cameras in my hand or any other place. I have not also seen any technical online video in detail about them . I know no detail about the cameras. I only suggested my idea through the theoretical scope. I discriminated between 3D cameras and 2D cameras. I had seen a video from Boston Dynamics Atlas robot and also in TV, in which some cameras render the environment image in 3D scan and can analyze the environment for collision avoidance. A 3D scanner camera is suitable for advanced motion planning. But I guess some simpler cameras you could possibly use for your case. I mean a camera which can distinguish between the rough terrain where the farms are bordered, and the traction where the crops are grown. This could be simply available by a camera which could detect two different colors: color of row crop, and the curb obstacles along borders of any farm. There might be no need for an advanced 3D scanner camera, so long as a simple camera can distinguish dark and light fields and tractions over the farming land based on their color (grey and black). This is all I knew.
  • asked a question related to Autonomous Vehicles
Question
18 answers
Will the development of autonomous cars be correlated with the development of electric cars?
Will these technologies be developed in parallel?
In the future, will a significant part of autonomous cars also be electric cars?
Please reply
I invite you to the discussion
Thank you very much
Best wishes
Relevant answer
Answer
Dear James Marson,
Thanks for the positive assessment of this discussion on the development of autonomous motor vehicle technology in the potential synergy of development with electromobility. Thanks for your comment. I agree with your view on this topic.
Regards,
Dariusz Prokopowicz
  • asked a question related to Autonomous Vehicles
Question
2 answers
I conduct the lane changing decision-making model based on stackelberg game for autonomou vehicles. However, the stackelberg game is bi-level optimal problem show as below appedix file and how to attain the stackelberg equilibrium solution that distorb me for a long time. I use SQP methond in MATLAB to get the stackelberg game optimal, but the results is terribled. Can I use SQP methon to get the soluton? And could you please indict the thinking for solve for me to solve. Thank you very much.
  • asked a question related to Autonomous Vehicles
Question
2 answers
I have been looking into self-supervised methods in computer vision. For example which looks at three consecutive frames of video, tasks a network with predicting the third frame, and uses the original as the ground truth / supervisory signal.
This type of pretext task for self supervision is a cross between context-based and semantic label based pretext tasks
Lane line detection in many ways is approached as a subset of the semantic segmentation task.
I am wondering if there is any way to come up with a pretext task that is specific to lane line detection?
I have seen where self supervision is used in the lane fitting task
But this is used after the lane segments have been identified.
Relevant answer
Did you mean to type pre-test?
  • asked a question related to Autonomous Vehicles
Question
2 answers
I am working on an autonomous driving domain and my field poses multiple autonomous vehicle collision avoidances. I want to try to solve it using RL. In this case, the following statements regarding the RL method are correct or any thins complicated with the fundamentals of RL. Please explain to me.
"Through Learning Control, control knowledge of a control function can be created through the training by Reinforcement Learning. However, the conventional Reinforcement Learning method does not provide the application of more than one control function within a Learning Control System. Execution of more control functions within a Learning Control System would require the application of multiple learning processes within a control system. Methods concerning the application of learning processes in Learning Control vary depending on the application of the control device and the purpose of the system."
Relevant answer
Answer
The problem under discussion is obstacle avoidance by reinforcement learning. Reinforcement learning works based on the concept of reward based action. So, every time a obstacle is detected, your logic should be such that advancement or activation happens on a particular front which could be treated as the reward. Hope this helps a bit.
  • asked a question related to Autonomous Vehicles
Question
10 answers
With the Trolley Problem being a hot disucssion with lots of phylosophical and technical papers discussed on it. It is eventually very hard to get to a final idea whether to incorporate this problem into AVs driving ethics? do they really create the dilemma ?
in 2017, the German ethics commission for automated and connected driving released 20 ethical guidelines for autonomous vehicles.
and Yet one more question : Why this comission paper, have so few references? does it mean they have cared less about other phylosophers ideas? or they have tried to found some ideas?
There are lots of relevant papers. bring them in this post to discuss it.
Relevant answer
Yes I agree, there are bad humans who drive cars, design algorithms and write code, along with many other things. I assume you are not suggesting that robot makers are a cut above the rest of humanity and can make these determinations and dispense these permissions for the purpose of protecting the rest of us from bad drivers because we all hate (no argument there) bad drivers. How about killer robots and the authors of such protocols; call to:{contingent factors when it is necessary to choose whose death}
Is this an omelette making egg breaking argument because if so I find they tend to end by generalising about the quality of human types and their utility to other human types. Not a good outcome from my reading of history but who knows what the future may bring. Maybe we could build a robot better than human and all mass suicide for being slower dimmer lesser creatures. Who’d argue with that?
  • asked a question related to Autonomous Vehicles
Question
4 answers
Autonomous vehicle, robots and rovers are finding applications in space exploration. Which tasks can Autonomous rovers and robots can take over from human (In other planets like mars, or moon or titan etc). How can an army of robots create all the infrastructure that humans needs for settlement?
Relevant answer
Answer
I would call it a network of autonomous rover. Their instruments will perform with a
better resolution and will be more versatile.
Will include deeper drilling, moving radar, etc. Also they will communicate with each other.
  • asked a question related to Autonomous Vehicles
Question
5 answers
Hello
I want to find accurate locations of some static objects to be used as landmarks for localizing other mobile object. As the landmarks will be used to localize other object, their own location must be accurate. Kindly suggest some practical methods for accurate positioning of static objects(landmark).
Best regards
Relevant answer
Answer
@tariq islam .Use dead reckoning.it is the process of calculating current position of some moving object by using a previously determined position, or fix, and then incorporating estimations of speed, heading direction, and course over elapsed time.
  • asked a question related to Autonomous Vehicles
Question
3 answers
I have attached two papers, which discuss the trolley problem differently. one trying to follow the question and picture possible cases, the other, "Avoids" the unfairness of the Trolley problem, but its discussion is not long enough and gets to a conclusion fast. Which doctorine of thought do you prefer?
Relevant answer
Answer
I think both references target different problems altogether and it is not a matter of choosing doctrines. [1] is aimed at the problem that " the thesis that killing is worse than letting die cannot be used in any simple, mechanical way in order to yield conclusions about abortion, euthanasia, and the distribution of scarce medical resources." meanwhile [2] is aimed at autonomous vehicles.
Even putting this difference aside [2] states that "It is important to focus not on abstract thought experiments but on concrete conditions that influence the behavior and properties of self-driving cars as being developed through
different stages lading to deployment and inclusion in traffic". This also separates the scope of work of [2] with respect to [1].
I think [2] does not address the Trolley Problem and merely evades it. The statement "The relevant trolley problem scenario in the context of the state
of the art technology is thus more likely to be: hitting an obstacle that is correctly identified versus hitting another object that is unknown or incorrectly identified. It is unrealistic to assume that the self-driving car will have information about whether a human dies or not in a critical situation" is a different problem altogether. There are instances of the problem that take into account this issues and still the Trolley problem appears. further the stament in [2] that "When neural networks are trained by analyzing photos of pedestrians, are those photos subject to fairness?" is completely different than the Trolley problem.
In my sincere opinion I think that [2] should rename the title of the publication since it is a different set of problems that do not address the trolley problem, but are very important to take into account for AV design.
[1]Thomson, J. J. (1976). Killing, letting die, and the trolley problem. The Monist, 59(2), 204-217.
[2] Holstein, T., & Dodig-Crnkovic, G. (2018, May). Avoiding the intrinsic unfairness of the trolley problem. In Proceedings of the International Workshop on Software Fairness (pp. 32-37).
  • asked a question related to Autonomous Vehicles
Question
5 answers
Anyone can suggest a dataset of scenarios/test cases that could be used to test and validate autonomous vehicles? I am interested in scenarios that are described in english or in any scenario description language that could be used for virtual closed loop testing?
Relevant answer
Answer
  • asked a question related to Autonomous Vehicles
Question
4 answers
Hi All
I'm looking for some papers on Self driving cars that discuss the current capability and work, on how much Data processing these cars do, and what are the steps and techniques applied to improve them
Relevant answer
Answer
@Farzad possibly this could help your cause.
Yeshodara, N. S., Nagojappa, N. S., & Kishore, N. (2014, October). Cloud based self driving cars. In 2014 IEEE International Conference on Cloud Computing in Emerging Markets (CCEM) (pp. 1-7). IEEE.
  • asked a question related to Autonomous Vehicles
Question
7 answers
We want to design a RL framework to help the autonomous vehicle decide when to lane change and how to perform the movements( the lateral and longitudinal acceleration). Can a flat DQN output the action (0,1) , 1 represent conduct lane change now, 0 represent not to conduct lane change now, while also output the action for continuous lane change movements?
Relevant answer
Answer
U can use a Q function to learn values of making and not making a lane change and choose greedily. But for contious action space, Deep Q Learning is not preferred (but could be done, u need to predect mean and standard deviation and then sample), instead architecture like PPO is better for continious control
  • asked a question related to Autonomous Vehicles
Question
9 answers
Hello
I need to know about the ways to represent a path mathematically and technically in the vehicle industry based on a map which contains lanes. An initial search shows that path generally defined sometimes using some way-points, others use curvilinear coordinates and others define a network of connected segments. Which are the relevant and common ways to represent a path on map with lanes ? Any citations about the matter ?!
Cordially
Relevant answer
Answer
Hi,
This is a very interesting research domain.
Here are some sources:
J. Kong, M. Pfeiffer, G. Schildbach and F. Borrelli, "Kinematic and dynamic vehicle models for autonomous driving control design," 2015 IEEE Intelligent Vehicles Symposium (IV), Seoul, Korea (South), 2015, pp. 1094-1099, doi: 10.1109/IVS.2015.7225830.
Good luck with your studies.
  • asked a question related to Autonomous Vehicles
Question
4 answers
Hi there, im currently writing my dissertation on how autonomous vehicles will influence real estate valuation.
To define real estate value factors (factors that residents of city X find influential in determining residential valuation) I am conducting a qualitative survey to exhibit what people are most attracted to.
How can I justify my survey size, of say 100 people?
  • asked a question related to Autonomous Vehicles
Question
6 answers
Colleagues and I are about to begin a project investigating the role of town planners and local authorities in establishing the introduction of connected and autonomous vehicles (CAVs).
What views do you have on the needs of users of CAVs, particularly members of the disabled community; the usefulness of planners creating infrastructure to accommodate CAVs and join these with existing conventional forms of travel; and how these can combine to establish a 'whole of journey' approach to future transport systems?
Relevant answer
Answer
Dear Martin Schmidt many thanks for the link - very helpful for our work.
Very best wishes, James.
  • asked a question related to Autonomous Vehicles
Question
7 answers
Dear Researchers,
I am pursuing my masters in Mechatronics (Germany) and currently working as a system validation intern. I am looking for some advice/guidance on possible topics in the field of autonomous driving. I am open to any sub-field but my current experience is in system validation of ADAS/ Autonomous Driving mostly in a Software-in-Loop environment. There is currently not a industry accepted standard for validation of ADAS/AD features so something along these lines would be greatly appreciated
Thanks in advance!
Relevant answer
Answer
Hello All,
I am searching for topics for master's thesis in the areas of ADAS, IoT.
Could someone please suggest my cutting-edge topics?
in case of any queries please mail me on qutibahhussein862@gmail.com
Thank you in advance
Best Regards,
Qutibah Hussein
  • asked a question related to Autonomous Vehicles
Question
6 answers
hi, now I am researching about these things, this is my first time to ask question at this site. I need various opinions who is expert about this topic. Please help me !!
How do you think the social environment will change in the near future when full-autonomous driving becomes the norm?
1. Future trends in the automobile market
2. Trends in technology announcements including automobiles
3.Changes in the living environment of general consumers
*English, Chinese OK:)
Relevant answer
Answer
Hi - in terms of the changes in the living environment i think the opening of access to convenient and perhaps personal means’ of transport for people with disabilities will be hugely significant in the introduction of CAVs. Also, given the move to electric vehicles, which most CAVs are likely to be, the benefits to clean air initiatives will also be important in peoples lives and their exposure to toxins in highly traffic-dense areas.
Hope this helps and very best wishes, James.
  • asked a question related to Autonomous Vehicles
Question
40 answers
In my opinion, the development of the necessary infrastructure and security stabilities is of key importance for the development of autonomous cars technology, so that the development of autonomous cars technology and the increase in the number of autonomous cars does not increase statistics on the number of road accidents.
In addition, the development of autonomous cars technology can be paralleled to the development of electromobility. For the development of electromobility and the number of used electric cars, it is also necessary to build the necessary infrastructure installed on roads, urban streets and interurban arteries of communication charging points for batteries into electricity.
In some countries there are active policies for the development of electromobility, under which the state from public finance funds pays extra to purchase an electric car and invests in projects to develop the necessary infrastructure for charging points in electricity. Other power plants are also being built as part of the development of renewable energy sources, because the development of electromobility is causing a significant increase in electricity demand. Unfortunately, this pro-ecological, active policy for the development of electromobility is carried out only in some countries.
Do you agree with my opinion on this matter?
In view of the above, I am asking you the following question:
What are the main determinants of the development of electromobility and autonomous cars technology?
Please reply
I invite you to the discussion
Thank you very much
Best wishes
Relevant answer
Answer
I a gree with Dariusz Prokopowicz
  • asked a question related to Autonomous Vehicles
Question
5 answers
Hi,
I am designing a model for detecting attacks launched against connected vehicles (autonomous vehicles). Where can I find a dataset having connected vehicle or connected vehicle network related attacks?
Relevant answer
Answer
  • asked a question related to Autonomous Vehicles
Question
3 answers
A monocular camera is to be calibrated, which is located in the area of the vehicle and looks in front of the direction of travel. During the calibration, the extrinsic parameters (position and rotation between the camera coordinate system and the origin of the vehicle coordinate system: Center of the rear axle of the vehicle) should be calculated. The camera parameters are calculated online, i.e. while the camera is taking pictures. The algorithm should automatically calculate the extrinsic parameters from driving scene images while driving.
I am looking forward to your feedback !
Relevant answer
Answer
you can take a video stream on a straight line and then use motion vectors
  • asked a question related to Autonomous Vehicles
Question
20 answers
Do you think that artificial intelligence will be implemented in the control systems of driving and orientation in the field in autonomous cars?
What are the effects of artificial intelligence implemented in the field of driving control systems and orientation in the field of autonomous cars?
Will autonomous cars be safe?
Will autonomous cars be mostly electric cars at the same time?
Please, answer, comments.
I invite you to the discussion.
Best wishes
Relevant answer
Answer
Thank you for your response. I agree with you. You added some very interesting information to our discussion. I also believe that the importance of artificial intelligence is growing in the implementation of this technology in various applications, in various fields, branches and sectors of the economy and in the improvement of research works. Thank you very much for proposing an article on this important issue artificial intelligence.
Greetings, Have a nice day,
Dariusz Prokopowicz
  • asked a question related to Autonomous Vehicles
Question
1 answer
Hi,
Kindly can anyone suggest a multiagent based traffic simulator for selecting of best possible routes?
Regards.
Relevant answer
Answer
Hello Anum.
There are a plenty of agent-based traffic simulators. I used SUMO (https://www.eclipse.org/sumo) a couple of years ago, which is a powerful microscopic traffic simulator. I also recommend you this book of J. Barceló (https://www.springer.com/gp/book/9781441961419). It presents the basic of traffic simulation and gives details of many simulators.
On the other hand, there are some general purpose agent-based simulators. Many people use them to model and simulate traffic scenarios. Examples include NetLogo (https://ccl.northwestern.edu/netlogo) and Repast (https://repast.github.io).
Cheers,
Marcelo
  • asked a question related to Autonomous Vehicles
Question
3 answers
Dear researchers,
if you are you interested in research on efficient autonomous vehicles, I have something new for you.
We have just launched our new open source reinforcement learning environment. Here you can find it: https://github.com/dynamik1703/gym_longicontrol
Our new environment is in the field of autonomous driving. It offers the possibility to test and further develop algorithms for the efficient longitudinal control.
The longitudinal control problem has various challenges. One example is the trade-off between conflicting goals of travel time minimization and energy consumption. They contradict each other because a fast driving vehicle leads to high-energy consumption and vice versa.
Through the proposed RL environment, which is adapted to the OpenAi Gym standardization, we show that it is easy to prototype and implement state-of-art RL algorithms. Besides, the LongiControl environment is suitable for various examinations. In addition to the comparison of RL algorithms and the evaluation of safety algorithms, investigations in the area of Multi-Objective Reinforcement Learning are also possible. Further possible research objectives are the comparison with planning algorithms for known routes, investigation of the influence of model uncertainties and the consideration of very long-term objectives like arriving at a specific time.
LongiControl is designed to enable the community to leverage the latest strategies of reinforcement learning to address a real-world and high-impact problem in the field of autonomous driving.
Have fun trying it out! If you have any questions, feel free to write.
  • asked a question related to Autonomous Vehicles
Question
2 answers
Hi,
I built a Simulink vehicle model with Dugoff's tire model that requires two parameters to calculate Fx and Fy; namely, the longitudinal stiffness and the cornering stiffness. Now I would like to verify my Simulink model by comparing the results with MSC ADAMS. ADAMS does not have Dugoff's tire model. How can I get these two parameters from ADAMS ?
Relevant answer
Answer
you can validate your data by using Dugoff's tire model which you made in matlab. in other world you should use adams matlab co-simulation. first made tire model in matlab. Then use this model as adams input. then run adams model.(your adams model with matlab tire model.)finally compare the adams vehicle model with matlab vehicle model.
Furthermore, you can use magic formula pacajka model to compare .It is more accurate.
  • asked a question related to Autonomous Vehicles
Question
1 answer
I was told by a person without any reference,
Co-60 source mus be replaced after the dose rate reached 50 cGy/minute..please anyone may guide me to provide the correct reference??
Relevant answer
Answer
is there any reference about replacement of source after two half lives.
  • asked a question related to Autonomous Vehicles
Question
5 answers
I some papers, the mean square error is considered. In some other, the mse is normalized by dividing the error by the (total sampling instants x the total length of the reference trajectory).
Which solution does represent the factual error?
Relevant answer
Answer
Dear Samira Eshghi,
The performance of the controller is generally measured using the following parameters such as :
1) Maximum Overshoot
2) Settling Time
3) Rise Time
4) Steady State Error.
Additionally, there are other ways to measure performance in literature. This includes integrated absolute error (IAE), the integral of squared-error
(ISE), or the integrated of time-weighted-squared-error (ITSE). These performance measures have their own advantages and disadvantages. For example, the disadvantage of the IAE and ISE criteria is that its minimization can result in a response with relatively small overshoot but a long settling time because the ISE performance criterion weights all errors equally independent of time. Although the ITSE performance criterion can overcome the disadvantage of the ISE criterion, the derivation processes of the analytical formula are complex and time-consuming. The formulas for these performance measures or the performance measure based on overshoot, settling time, rise time and steady-state error can be found in the paper given below.
In discrete domain, ISE refers to Mean Square Error. I am unfamiliar with normalized-mean-square-error .
  • asked a question related to Autonomous Vehicles
Question
5 answers
The challenge faced by the autonomous vehicle industry is to keep pace with the exploration of autonomous vehicles. The industrial research centers are limited and prevent them from sharing their thoughts. This in turn prevents their researchers from gaining access to other knowledge pools.
It is important for industries to come together and maintain an open relationship with various research universities in order to find the best solution for the development and evaluation of autonomous vehicles so that autonomous vehicles meet the basic safety and protection criteria.
Do we have a specific model and strategy for industry-university collaboration on autonomous vehicle products and services?
Relevant answer
Nice Dear John Mendy
  • asked a question related to Autonomous Vehicles
Question
4 answers
I am working on Application - Specific Radio Resource Allocation in 5G and I need to know the size of packets used for every application or use-case of 5G if possible. Applications like; Video conferencing, AR/VR, Web browsing, Wireless E-health, Industry automation, Autonomous vehicles, Smart Energy, Smart City and so on.
Relevant answer
Answer
Thank you very much Sai R N.I am most grateful.
  • asked a question related to Autonomous Vehicles
Question
14 answers
It is widely believed that motor drive has reached a competent level of maturity where hardly gives birth to new innovations. Would it be possible to see a revival of theoretical and/or practical innovations in this field, in a foreseeable future, taking into account the rapid progress in microelectronics, power electronics, permanent magnet materials in one hand, and demanding new applications like autonomous vehicles, drones, and fly by wire on the other?
Relevant answer
It seems that the paucity of innovations in the field of electric motors, especially in recent years, has become a big concern for the researchers. To the best of my knowledge, this concern has become rampant to a great extent, however, I think vice versa. New types of electric motors are developed by a lot of researchers for a broad variety of industrial applications with a focus on electric vehicles, including VFRMs, VRMs, FSPMs, Memory machines, and etc, all of which are proposed to increase efficiency, power and torque densities, and decrease the consumption of rare-earth PMs. Innovation Continues. The following paper is a good source to support this opinion.
  • asked a question related to Autonomous Vehicles
Question
17 answers
Electric cars and hydrogen cars both run on apparently clean fuels. However, the source of electricity or hydrogen production may be based over partially clean processes. It is a question of the future, which of them will prevail.
Relevant answer
Answer
  • asked a question related to Autonomous Vehicles
Question
4 answers
I have designed a trajectory tracking controller for a passenger car. The output of the controller is the applied torque to the wheels. I decided to verify this controller using an IPG carmaker simulator. The operator of the IPG carmaker asked me to give the position of the accelerator and brake pedals as the control input; otherwise, he cannot perform the simulation. Is he right? Is not there any way to implement the desired torque to the wheels in IPG carmaker? If so, how can I convert the applied torque to the position of the accelerator and brake pedals? Is there any transfer function for this purpose?
Relevant answer
Answer
The wheel torque can be interpreted as throttle opening percentage by modeling the driveline backwards, i.e. wheels=> driveshaft=> gearbox=> move-off element (clutch or torque converter)=> engine, and finally the engine map can be used to obtain the throttle opening percentage. A similar approach for brake system can be explored as well.
I know it seems obvious, but you should perform this procedure using the parameters defined in your validation platform (IPG). Otherwise, you should find the detailed specifications of your considered vehicle.
  • asked a question related to Autonomous Vehicles
Question
4 answers
I,m looking for a research topic on pedestrian safety at intersections with the advent of autonomous vehicles. I'll appreciate any suggestions
Relevant answer
Answer
I suggest to start literature review related to your objectives.
Read the last 5 to 10 years most related published research works closer to your topic. In particular carefully read Conclusion and Future work section of each of these (5 to 10 years published work) papers, where new research directions proposals or open problems will be available for you.
Best of luck for your literature review!!!
  • asked a question related to Autonomous Vehicles
Question
2 answers
It is seen in the recent time in the news across the globe relevant to develop and deploy the autonomous vehicles on road as soon as earliest. Although, it is a step-by-step process in which we have to spend billions of dollars to develop fully autonomous vehicles. And even after the development of such FAVs, still, there are some questions remain open such as (a) What about the efficiency of the FAVs? (b) Who is the owner of the vehicle? (c) What about ethical and moral issues? and so on. In these circumstances, is it more useful to concentrate on DATs rather than FAVs?
Please share your viable opinion. 
Thank you, Indeed. 
Relevant answer
Answer
Ankit R. Patel I think the developers also sometimes underestimate peoples wishes and will to own and to control a vehicle. There is a deep feeling of freedom in driving a vehicle by your selves. The autonomous vehicle will probably be safer for the driver and the surroundings, but that has never stopped humans from wanting their own control.
  • asked a question related to Autonomous Vehicles
Question
24 answers
The new highway, expected to be one of many, will run between Beijing and the Xiongan New Area in the Hebei province, FutureCar reports. There, a new 62-mile stretch of freeway will have two lanes dedicated to autonomous vehicles (AVs). The idea is that the infrastructure investment will give AVs access to real-world traffic conditions — but also that the separate lanes will ensure that the still-limited AV tech is tested in a way that minimizes risk for human drivers.
Will these dedicated lanes do good or harm to the safety or mobility efficiency of the freeway?
Relevant answer
Answer
Seems like a good idea. If the operation of the AVs is hugely successful, they could be integrated with the general vehicles. If it's not successful at all, the lanes could be converted to general vehicle lanes at little cost. In the process we will all increase our knowledge of the risk associated with the operation of AVs.
  • asked a question related to Autonomous Vehicles
Question
7 answers
I'm looking for different concepts for determining a position of an autonomous vehicle. While researching different GPS based concepts I came across dGPS and RTK. I understand that RTK uses the carrier wave itself to determine the range between the satellite and the receiver which has a higher frequency than the code modulated on it. As far as I understand by using GPS the receiver matches his internal pseudo code to the one he receives. After matching the codes, the receiver reads the message sent. Doing that to 3 more satellites the receiver can calculate the time it took the signals to arrive and triangulate the receivers position.
RTK uses the carrier wave instead. By solving the integer ambiguity problem. The range to the satellite can be calculated by multiplying the wavelength with the number of cycles to it.
Both ways seem to me like they are just different ways to calculate the distance but apparently there is an advantage of using the carrier wave, as shown in the images under this link:
Could someone explain to me, why there is a time uncertainty? For me either there is a 1 or a 0, nothing in between..
Many thanks in advance.
Relevant answer
Answer
Dear Victor Egold, the reason why RTK is more accurate than DGPS is because of the method of the range measurement. RTK uses carrier phase-shift measurement while DGPS uses code measurement. In carrier phase measurement, the integer ambiguity is resolved. That is, the fractional part of the cycle in the signal is resolved. This resolved fractional part of the cycle is added to the total number of complete cycles before multiplying it with the wavelength to obtain accurate/precise range from the receiver to the satellite. The ambiguity is determined using three techniques: Single Differencing, Double Differencing and Triple Differencing. All of these techniques require that additional observations be obtained.
  • asked a question related to Autonomous Vehicles
Question
3 answers
Autonomous driving is a goal towards which the Automotive Industry is evolving in 5 steps now commonly recognised. Autonomous Driving is discussed under acronyms such as ADAS or CAV (connected and the autonomous vehicle).
However, is it leading ultimately towards more autonomy or more centralised supervision?
Relevant answer
Answer
Fully autonomous driving is still an area of research. However successful implementation of various ADAS modules and and invent of advanced communication and signal processing technologies used is Connected Vehicles can make it a reality in near future. To relate ADAS features to automation level, the U.S. department of transportation (DOT) has recently released a guideline for safe testing and deployment of automated vehicles. DOT gives a “Levels of Automation” classification for driverless cars as a guideline in SAE international [1].
The embedded processors within the vehicles are interacting constantly with environment and gathering enormous amount of data which in tern may lead to a centralized level supervision for the vehicles.
[1] SAE On-Road Automated Vehicle Standards Committee. "Taxonomy and definitions for terms related to on-road motor vehicle automated driving systems." SAE International (2014).
  • asked a question related to Autonomous Vehicles
Question
5 answers
In "Simultaneous Stabilization and Tracking of Nonholonomic Mobile Robots: A Lyapunov-Based Approach", the authors stated that the nonholonomy of mobile robots allows control of the system with less control inputs. A mobile robot is underactuated in nature, and it's often controlled with less number of control inputs than the system's degree of freedom (if dynamic is ignored). How does the nonholonomy of the system helps with that?
Relevant answer
Answer
Dear Samira Eshghi,
I suggest you to see links and attached files on topic.
Feedback Control of a Nonholonomic Car-like Robot - LAAS
Exponential stabilization of mobile robots with nonholonomic constraints
Control of a Nonholonomic Mobile Robot: Backstepping ... - CiteSeerX
Path planning and control of non-holonomic mobile robots
Best regards
  • asked a question related to Autonomous Vehicles
Question
17 answers
Traditionally the driving license is issued to adults 18 years of age or older, who pass the theoretical and practical test of the issuing authorities.
However, what about if the self driven cars become more available, AND the safety for the drivers and the pedestrians become even better than the current practices...
If such High Tech / AI supported cars are there, then would you think teenagers should be allowed to operate them by their own, with their guardians approval?
If yes, what is the lowest age limit you may consider to let them get a driving license for such a future vehicle? 14 year-old? 16? 12?
Why?
Relevant answer
Answer
In any case (including High Tech / AI supported cars), I believe that the minimum age to allow getting a driving license should be 18 years old.
  • asked a question related to Autonomous Vehicles
Question
6 answers
I know that in some years SAE Level 5 Autonomous Cars will be sold in the market, but probably just as concept cars. When do you think they will become a competitive solution in the market at affordable price?
Relevant answer
Answer
I think it will take long time (may be more than 10-20 years), as there are number of factors affecting/not clear yet, it includes:
1. Government policies
2. Acceptance and Trust level by the common people
3. Rules for the autonomy, specifically who will responsible for any accident, if any
4. Cyber Security threats
5. Road conditions
Thank you.
Ankit Patel
  • asked a question related to Autonomous Vehicles
Question
6 answers
will the autonomous vehicle replace the pleasure of driving??
Relevant answer
The autonomous driver cars can not completely substitute the manually driven cars. This may be an ultimate goal. Hybrid driven cars may be more realistic goal for the next time.
Best wishes
  • asked a question related to Autonomous Vehicles
Question
4 answers
In your opinion, what are the main challenges currently for the design of autonomous vehicle systems? Answers can be related to hardware, software, controls, societal impact, etc.
Secondarily, what are the main considerations that can be relaxed or approximated (to save design time and cost) for systems that will not directly interface with humans?
Relevant answer
Answer
I conclude that the idea that one can transfer control of the almost autonomous vehicle to the human occupant when the going gets too complicated for the system is fatally flawed. Driving skill assumptions about the hapless occupant are bound to be inaccurate and far lower than a well designed electronic system. If the situation is starkly perilous, the "operator" may well have fainted dead away. The closer the development process gets to the completely autonomous (level 5) design, the more insurmountable the handover. Even a moderately well designed level 5 system will be able to outperform 99% of the anticipated human operators. The user experience in the autonomous vehicle will, of necessity, be that of a train ride with no user concern for the technical or practical matters of the vehicle. The user experience in the almost autonomous (level 4.8 or so) vehicle may be so disconcerting as to prevent anyone who is informed of the control transfer situation from acquiring such a vehicle, or taking a ride taxi style.
John M. Kirkwood, CHFP
2208 Hunter Street Cinnaminson, NJ 08077
Phone: 609-828-2045
  • asked a question related to Autonomous Vehicles
Question
6 answers
Hello to all
Artificial intelligence is often mentioned as an area where corporations make large investments. It seems that they use AI in autonomous vehicles, face recognition, medicine, banking, financing, weather forecast, etc.
The celebrated backpropagation algorithm has wide use within academic circles. Many papers exist that discuss, analyze and apply BP. But is it actually used in any of the practical developments undertaken by large companies? If so, in which specific ones. If not, which is the preferred NN training method for these commercially oriented applications?
With cordial regards,
Daniel Crespin
Relevant answer
Answer
Thanks very much
The paper by Yann LeCun pointed to by Eduardo sufficiently answers my question. It is now clear that BP has not been replaced and remains important down to practical real world applications. From LeCun paper and after some navigation I found the following additional references:
Ian Goodfellow, I. et al Deep Learning (book by MIT Press)
Jürgen Schmidhuber, Deep Learning in Neural Networks. An Overview
Cireçan, D. et al - Flexible, High Performance Convolutional Neural Networks for Image Classification
Time ago I published the paper Generalized Backpropagation containing a deduction of general BP formulas
and with BP still useful may now write a further generalization.
Hearty thanks to all, especially to Eduardo.
Daniel Crespin
  • asked a question related to Autonomous Vehicles
Question
4 answers
We are in the planning stage we want to make a self driving vehicle, maybe anybody could point us to the right company or direction.
Relevant answer
Answer
I am doing research, maybe i can become the part of company for research direction and collaborate in research field, if you are interested, we can talk more through this channel, my email id
  • asked a question related to Autonomous Vehicles
Question
3 answers
In particular, how should autonomus vehicles be programmed to behave in the fat man trolley problem? An algorithm can instruct the machine to throw the fat man off the bridge?
Relevant answer
Answer
  • asked a question related to Autonomous Vehicles
Question
5 answers
Hi everybody,
I have a question about the circle of friction or Kamm circle. This circle is used in vehicle modelling, but I did not understand this well. I say my own opinion, please let me know if I am mistaken.
vehicle friction force at tire contact patch divided into two forces, lateral and longitudinal forces. As the coefficient of friction is u and the normal force is N, then the maximum of friction force is F_tot = uN. Now F_tot is the radius of the Kamm circle (?), then F_x and F_y are longitudinal and lateral forces, so ( F_x^2 + F_y^2 )^0.5 is lower or equal to F_tot.
If I was right, please tell me what is the use of Kamm circle?
Thanks in advance.
Relevant answer
Answer
Friction circle (based on the assumption that a tire can develop equal limiting longitudinal and lateral forces) gives you:
- The maximum longitudinal force that the tire can develop while it is developing a specified lateral force, or
- The minimum lateral force that the tire can developed while it is developing a specified longitudinal force.
Consequently, the maximum longitudinal force in the absence of lateral force generation and the maximum lateral force in the absence of longitudinal force will both be equal to adhesion coefficient*normal tire load.
For tires developed for emphasized traction or cornering capability, "friction ellipse" may be more suitable.
  • asked a question related to Autonomous Vehicles
Question
1 answer
Hi everyone,
I stock in the vehicle wheel's camber modelling.
I want the main ( non-linear ) equation between vehicle's wheel camber angle and side slip angle.
The linear equation is in the form below:
alpha_star = alpha + K_camb * gamma
The book in which the linear equation is available states that the main equation is available in the following references, BUT I can't find these references...
1 - M. Mitschke and H. Wallentowitz. Dynamik der Kraftfahrzeuge. Springer Verlag, Berlin, 4. edition, 2004.
2- J. Reimpell and P. Sponagel. Fahrwerktechnik: Reifen und Räder. Vogel Fachbuch. Würzburg, 1995.
Thanks for your patience and cooperation.
Relevant answer
Answer
Dear Arash,
Please take a look at links and attached files in topic.
-Vehicle Dynamics of Modern Passenger Cars
-Camber Angle Control Method Corresponding to the Electric Vehicle ...
-Camber-Car Design & Dynamics - Zonagravedad
-)) (( tan)/() )( 1( )) ( tan sin( S B BE S E S B C D F + + + −=Φ +Φ = α α γ ...
-performance et comportement des vehicules - ULg - LTAS
Best regards
  • asked a question related to Autonomous Vehicles
Question
3 answers
Good day,
I am doing my PhD in operation research in logistics. I want to create an urban logistics simulation, during which I would show how autonomous vehicles can adapt to disruptions and resilience would emerge. The theoretical approach is based on complex-adaptive systems theory.
As I understand, the routing should be made by using “reinforcement learning for combinatorial optimization”, however, I do not have experience related to RL application for route scheduling. Could anyone recommend courses or literature related to this topic?
If anyone has developed such a model, maybe could provide some insights in to the data architecture? As I understand this would be similar to supervised learning, however sequence of categories must also be taken into consideration. Should the implementation be based on graphs? Or should I have features categorizing a trip with evaluation of the cost function?
In my case, I am having a e-commerce industry with product delivery to end-consumer, during the day I am generating traffic jams, which would block the routs. The algorithm should learn from the environment and select better routes automatically by considering the goal function.
Relevant answer
Answer
Dear Valentas,
It should be pointed out that the data structure technique belongs to the large domain of mathematical optimization, and combinatorial optimization is an element of this set. In addition, heuristics and metaheuristic techniques are part of this domain and are an extension for the resolution of multiobjective optimization. It is clear that these techniques can be used in a complementary manner or in hybridization.
For more information about this sbject, i suggest you to see links and attached file in topic.
How Autonomous Vehicles are Driving a Shift in Supply Chains
Solving Combinatorial Optimization Tasks by Reinforcement Learning ...
Logistics 2050 A Scenario Study - DHL
big data in logistics - DHL
Best regards
  • asked a question related to Autonomous Vehicles
Question
8 answers
I'm studing in control engineering and I want to work on autonomous vehicles.
recently I was searching a lot but I have no idea about latest challenge in this field.
as autonomous vehicles systems has Perception, Planning, Control, and Coordination fields, I prefer to work on Control.
So, if you have any idea about latest challenge in Autonomous vehicles control system, tell me.
thanks.
Relevant answer
Answer
Dear see these topics:
  1. Heavy Truck Cooperative Adaptive Cruise Control: Evaluation, Testing, and Stakeholder Engagement for Near Term Deployment.
  2. Identifying Autonomous Vehicle Technology Impacts on the Trucking Industry.
  3. Reactive Path Planning based on Swarm Behavior Methods.
  4. Localisation by Employing Stereo Cameras.
Regards
  • asked a question related to Autonomous Vehicles
Question
3 answers
I am hoping to do a pilot survey of about 10 to 15 people, before employing the main survey. The survey is on obtaining public preferences and attitudes towards different levels of autonomous vehicles.
I was just wondering what exactly should I consider or include in the pilot study plan.
Thanks
Relevant answer
Answer
Thank you very much for the links and files attached professor Lafifi.
  • asked a question related to Autonomous Vehicles
Question
5 answers
Dear all
I wish to know whether any one is working on Radar system for autonomous vehicles. I wish to know who is the supplier of the same. I need only a few systems. Can i get any system that can be integrated with arduino for obstacle detection.
And i am looking for a low cost solution.
Relevant answer
Answer
I suggest you check the affordable Texas Instruments Evaluation Module AWR1642BOOST which incorporates the 77- and 79-GHz FMCW TI-AWR1642, or his brother EVM system based on the TI-IWR1443 which provides elevation besides azimuth (bearing) data. Texas has some nice demos of these systems available online and, besides the software develop tools, provides browser-based GUIs for testing the system.
  • asked a question related to Autonomous Vehicles
Question
2 answers
Commercial, industrial, and military vehicles are required to safely navigate predetermine paths in degraded visual environments (i.e., fog, snow, heavy rain, dust, ect.). Various techniques based on sensor fusion (i.e., CCD cameras, RADAR, Laser sensors, ultrasonic sensors, ect.), Machine Learning/AI, human-aided operations and other are applied.
Relevant answer
Answer
To your point and as I think of autonomous or remote-controlled, the differentiation in my mind is having a human driver-in-the-loop for autonomous vehicles. Remotely controlled has access for “in situ” data and an observer to command maneuvers. However, the controls and actuation for an on-ramp entry maneuver into congested freeway, for example, is quite challenging and requires a driver-in-the loop for safe driving operations. As the environment or weather conditions could be classified as degraded, the aforementioned research problem continues to exist. I agree that range and range-rate information for targets in the field-of-view are reported are reported by RADAR; however please expand on sensors necessary for capturing and reporting image data and to provide insights about a degraded visual environment.
  • asked a question related to Autonomous Vehicles
Question
5 answers
Currently, there is plenty of literature available over user preferences or user perceptions towards autonomous vehicles. I am struggling to form a theme for my stated preference survey over levels of autonomous vehicles that is because the idea is not clear. Can anybody suggest me a theme over which I can form discrete choice experiments.
Present Levels of automation
Level 0 - No automation
Level 1 (Hands on) -Some automation modes are available.Driver and system shares the control of the vehicle. 
Level 2 (Hands off) -Some automation modes with full system control over vehicle. Driver monitors driving and should be prepared to intervene immediately. 
Level 3 (Eyes off ) - Full system control. Driver do not need to monitor driving and he can do other activities such as using phone or watch movie. Driver attention required in limited time specified by the manufacturer of car.
Future levels of automation:
Level 4 (Mind off) - Similar to level 3, but no driver attention is required for safety. Driver can even sleep and leave driving seat. Level 4 automation supported in limited areas only such as traffic jams. Outside these areas driver should take control.
Level 5 (Steering Wheel Optional) - Steering wheel is optional, no human intervention is required,i.e. A robotic car
Relevant answer