No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Robotics Use Case Scenarios
Abstract
Artificial Intelligence (AI) is widely used to accelerate state-of-the-art vision algorithms used in different domains including navigational algorithms, image segmentation, object classification, etc. Vision systems generate substantial amounts of data and it is desirable to run such AI algorithms at the edge where the data is being generated. Despite exhibiting higher accuracy than human beings, such AI algorithms must be accelerated in dedicated hardware. The Sundance's VCS-1 system, formerly T Agri Starter Kit, is the ideal platform to run the state-of-the-art AI algorithms. In this chapter, three use-cases are presented where the TAK and the T toolchain is being used to accelerate vision algorithms.
After making progression in developing the fundamental problems related to single-robot control, many researchers swerved and diverged their focus to studying multi-robot coordination. This research aims to take the motion planning and control (MPC) problem of a multi-robot system into a new space by considering a compartmentalized robot. An efficient variant of globally rigid formation, in which multiple car-like units are adjoint and move in parallel without collisions. The motion is governed by one of the sub-units acting as a leader, while other units maintain the fixed distance amongst each other and the leader in a rigid formation. The minimum distance technique is an important input to facilitate collision avoidance, robot decision making, and robot navigation. In this study a novel method to analytically compute the minimum distance between the closest point on the line segments of rectangular protective region and the obstacle is presented. Utilizing the Lyapunov-based Control Scheme a set of autonomous controllers are designed. Computer simulations of the proposed Lyapunov-based controllers for the compartmentalized robot are presented in interesting scenarios to show the efficacy of the unique set of controllers. In these simulations, the compartmentalized robot shows strict maintenance of a rigid formation with efficient collision and obstacle avoidance. The results open up research in the design and implementation of controllers by considering multiple compartmentalized robots into swarm models, splitting and re-joining units, and applying rotational leadership ideas.
Forestry and agricultural robotics are growing areas of research within the field of Robotics. Recent developments in planning, perception and mobility in unstructured outdoor environments have allowed the proliferation of innovative autonomous machines. The SEMFIRE project proposes a robotic system to support the removal of flammable material in forests, thus assisting in landscaping maintenance tasks and avoiding the deflagration of wildfires. In this work, we describe work in progress on the development of the Ranger, a large heavy-duty forestry machine based on the well-known Bobcat T190, which is the main actor of SEMFIRE. We present the design of the machine, which has been expanded with several sensors, its full integration in the Robot Operating System, as well as preliminary results.
Landscape maintenance is a pivotal factor to reduce wildfire hazard potential. Being a difficult, time-consuming and generally expensive task, humans often face great difficulties to address the problem effectively. Considering its spatial distribution and exploration capabilities, swarm robotics presents itself as a natural solution to assist humans or other robotic agents in the reconnaissance of forestry environments and identification of fuel accumulation, such as forest debris. In this work, we propose an optimization framework for the exploration problem of forestry environments with a team of swarming robots, which also entails the deployment of a network infrastructure formed by the robots to assist agents during fire-prevention tasks. Experiments show the benefits of employing the multi-robot coordination algorithm described, which provides an effective solution for the exploration task and guarantees a final formation that maximizes network coverage for mission information exchange, considering obstacles density and other interfering factors. Thus, the system proposed might assist human operators with relevant forestry information acquired during exploration and maintain a fail-safe network for communication while agents operate in the field.
Europe has been affected by an alarming number of wildfires every year, ravaging nearly half a million hectares of forestry areas and resulting in an unacceptable amount of human losses. The year 2017 was one of the worst ever recorded, with devastating wildfires raging across the globe. Bearing these shocking facts in mind, this position paper aims to lay the foundations of crucial new upcoming research in the field of forestry robotics by presenting an overview of the SEMFIRE project. SEMFIRE proposes the development of a robotic system designed to reduce the accumulation of combustible material in forests, thus assisting in landscaping maintenance tasks for wildfire prevention. This paper describes the multi-robot system envisaged by SEMFIRE, combined with pervasive local positioning, multi-robot navigation and coordination in forests, and an innovative artificial perception architecture.
This paper investigates the application of linear learning techniques to the place recognition problem. We present two learning methods, a supervised change prediction technique based on linear regression and an unsupervised change removal technique based on principal component analysis, and investigate how the performance of each is affected by the choice of training data. We show that the change prediction technique presented here succeeds only if it is provided with appropriate and adequate training data, which can be challenging for a mobile robotic system operating in an uncontrolled environment. In contrast, change removal can improve place recognition performance even when trained with as few as 100 samples. This paper shows that change removal can be combined with a number of different image descriptors and can improve performance across a range of different appearance conditions.
Forest fires affect Portugal and other Southern European countries to a greater or lesser extent every year, causing severe environmental, social and economic damage. This research is based upon the analysis of the results of a case study on the technical and social perceptions of multiple dimensions of forest fires in Portugal, which brings together the views and perspectives of different entities with expertise in forest management and in Forest Protection against Fires, forest owners and members of the public. Assessing perceptions of forest fires has proved to be relevant when it comes to a greater involvement of the local population in forest management. The results show that the opinions of local respondents were consensual in identifying crime as the main cause of the forest fires. Additionally, is the repeated failure to clean the forest biomass which cause the rapid spread of forest fires. Damage caused to the environment was more of an issue with the local forestry technicians and economic issues were at the forefront for the local community. Various solutions were put forward to mitigate these problems, but everyone agreed that cleaning the forests biomass and punishing arsonists should be at the top of the list of priorities
Technology has become the solution to many long-standing problems, and while current technologies may be effective, it is far from fully addressing the hug, complex, difficult and challenging tasks associated with disaster missions and risky intervention. The challenge is in finding creative, reliable and applicable technical solutions in such highly constrained and uncertain environment. In addition, it is necessary to overcome constrains on resources by developing innovative, cost effective and practical technology. Robotics can play important intelligent and technological roles that support first response equipment in harsh and dangerous environments while replacing rescue personnel from entering unreachable or unsafe places. Robotics solutions that are well adapted to local conditions of unstructured and unknown environment can greatly improve safety and security of personnel as well as work efficiency, productivity and flexibility. Solving and fulfilling the needs of such tasks presents challenges in robotic mechanical structure and mobility, sensors and sensor fusion, autonomous and semi autonomous control, planning and navigation, and machine intelligence. This paper categorizes the source of disasters and associated missions, and highlights the needs for suitable and reliable technology and technical and functional requirements of robotic systems to fulfill task objectives. In addition, it shows that robotic technologies can be used for disasters prevention or early warning, intervention and recovery efforts during disasters with all possible kinds of relevant missions while ensuring quality of service and safety of human beings. Some of these missions may include: demining, search and rescue, surveillance, reconnaissance and risk assessment, evacuation assistance, intrusion/victim detection and assessment, etc.
The DARPA PerceptOR program has implemented a rigorous evaluative test program which fosters the development of field relevant outdoor mobile robots. Autonomous ground vehicles were deployed on diverse test courses throughout the USA and quantitatively evaluated on such factors as autonomy level, waypoint acquisition, failure rate, speed, and communications bandwidth. Our efforts over the three year program have produced new approaches in planning, perception, localization, and control which have been driven by the quest for reliable operation in challenging environments. This paper focuses on some of the most unique aspects of the systems developed by the CMU PerceptOR team, the lessons learned during the effort, and the most immediate challenges that remain to be addressed.
The use of global positioning system (GPS) in outdoor localization
is quite a common solution in large environments where no other
reference is available and there are not so demanding positioning
requirements. Of course, fine motion without the use of an expensive
differential device is not an easy task, even now that available
precision has been greatly improved as the military encoding has been
removed. In this paper we present a localization algorithm based on
Kalman filtering that tries to fuse information coming from an
inexpensive single GPS with inertial data and map-based data. The
algorithm is able to produce an estimated configuration for the robot
that can be successfully fed back in a navigation system. Some
experiments show difficulties and possible solutions of this sensor
fusion problem
This paper presents MoDSeM, a software framework for cooperative perception supporting teams of robots. MoDSeM aims to provide a flexible semantic mapping framework able to represent all spatial information perceived in missions involving teams of robots, and to formalize the development of perception software, promoting the implementation of reusable modules that can fit varied team constitutions. We provide an overview of MoDSeM, and describe how it can be applied to multi-robot systems, discussing several sub-problems such as history and memory, or centralized vs distributed perception. Aiming to demonstrate the functionality of our prototype, preliminary experiments took place in simulation, using a m simulated map to demonstrate its ability to receive, store and retrieve information stored in semantic voxel grids, using ROS as a transport layer and OpenVDB as a grid storage mechanism. Results show the appropriateness of ROS and OpenVDB as a back-end for supporting the prototype, achieving a promising performance in all aspects of the task. Future developments will make use of these results to apply MoDSeM in realistic scenarios, including multi-robot indoor surveillance and precision forestry operations.
The ability to safely navigate is a crucial prerequisite for truly autonomous systems. A robot has to distinguish obstacles from traversable ground. Failing on this task can cause great damage or restrict the robots movement unnecessarily. Due to the security relevance of this problem, great effort is typically spent to design models for individual robots and sensors, and the complexity of such models is correlated to the complexity of the environment and the capabilities of the robot. We present a semi supervised learning approach, where the robot learns its traversability capabilities from a human operating it. From this partially and only positive labeled training data, our approach infers a model for the traversability analysis, thereby requiring very little manual effort for the human. In practical experiments we show that our method can be used for robots that need to reliably navigate on dirt roads as well as for robots that have very restricted traversability capabilities.
Forestry workers have previously been identified as an occupational group at high risk of work-related injury. This paper reviews factors affecting injury among forestry workers and draws on international literature in general (New Zealand studies in particular). A major proportion of the literature relates to tasks involving chain saws, such as felling and delimbing, and the operation of forestry machinery. By comparison, less attention has been given to tasks such as tree planting and pruning. And while physical risks have been well documented, comparatively little research has focused on mental workload, particularly in relation to motor-manual logging. Few studies have focused on the effects of work organization (e.g., hours of work, contracting) on injury rates.
In our paper we present an innovative locomotion concept for rough terrain based on six motorized wheels. Using rhombus configuration, the rover named Shrimp has a steering wheel in the front and the rear, and two wheels arranged on a bogie on each side. The front wheel has a spring suspension to guarantee optimal ground contact of all wheels at any time. The steering of the rover is realized by synchronizing the steering of the front and rear wheels and the speed difference of the bogie wheels. This allows for precision maneuvers and even turning on the spot with minimum slippage. The use of parallel articulations for the front wheel and the bogies enables to set a virtual center of rotation at the level of or below the wheel axis. This insures maximum stability and climbing abilities even for very low friction coefficients between the wheel and the ground.A well functioning prototype has been designed and manufactured. It shows excellent performance surpassing our expectations. The robot, measuring only about 60 cm in length and 20 cm in height, is able to passively overcome obstacles of up to two times its wheel diameter and can climb stairs with steps of over 20 cm.
Preliminary research on robotic vision in a regenerating forest environment <http://cfs.nrcan.gc.ca/publications?id=4582>
- FA Gougeon
- PH Kourtz
- M Strome
Gougeon, F.A., Kourtz, P.H., Strome, M.: Preliminary research on robotic
vision in a regenerating forest environment. In: Proc. Int. Symp. Intelligent
Robotics Systems, vol. 94, pp. 11-15 (1994). URL http://cfs.nrcan.gc.
ca/publications?id=4582
- J San-Miguel-Ayanz
- E Schulte
- G Schmuck
- A Camia
- P Strobl
- G Liberta
- C Giovando
- R Boca
- F Sedano
- P Kempeneers
- D Mcinerney
San-Miguel-Ayanz, J., Schulte, E., Schmuck, G., Camia, A., Strobl, P., Liberta, G., Giovando, C., Boca, R., Sedano, F., Kempeneers, P., McInerney, D.:
Comprehensive monitoring of wildfires in Europe:the European Forest Fire Information System (EFFIS),. Tech. rep., EuropeanCommission, Joint Research
Centre, Italy (2012)
- Xilinx
Xilinx: Vitis AI User Guide (2019).
URL https://www.xilinx.
Preliminary research on robotic vision in a regenerating forest environment
- F A Gougeon
- P H Kourtz
- M Strome
- FA Gougeon