Marionette: Enabling On-Road Wizard-of-Oz Autonomous Driving Studies

Conference Paper · March 2017with 340 Reads
DOI: 10.1145/2909824.3020256
Conference: the 2017 ACM/IEEE International Conference
Cite this publication
Abstract
There is a growing need to study the interactions between drivers and their increasingly autonomous vehicles. This paper describes a method of using a low-cost, portable, and versatile driver interaction system in commercial passenger vehicles to enable on-road partial and fully autonomous driving interaction studies. By conducting on-road Wizard-of-Oz studies in naturalistic settings, we can explore a range of driving conditions and scenarios far beyond what can be conducted in laboratory simulator environments. The Marionette system uses off-the-shelf components to cre- ate bidirectional communication between the driving controls of a Wizard-of-Oz vehicle operator and a driving study participant. It signals to the study participant what the car is doing and enables researchers to study participant intervention in driving activity. Mar- ionette is designed to be easily replicated for researchers studying partially autonomous driving interaction. This paper describes the design and evaluation of this system.

Do you want to read the rest of this conference paper?

Request full-text
Request Full-text Paper PDF
  • Thesis
    This thesis investigates the intersections of three disciplines, that are Design Research, Human-Robot Interaction studies, and Child Studies. In particular, this doctoral research is focused on two research questions, namely, what is (or might be) the role of design research in HRI? And, how to design acceptable and desirable child-robot play applications? The rst chapter introduces an overview of the mutual interest between robotics and design that is at the basis of the research. On the one hand, the interest of design toward robotics is documented through some exemplary projects from artists and designers that speculate on the human-robot coexistence condition. Vice versa, the robotics interest toward design is documented by referring to some tracks of robotic conferences, scienti c workshops and robotics journals which focused on the design-robotics relationship. Finally, a brief description of the background conditions that characterized this doctoral research are introduced, such as the fact of being a research founded by a company. The second chapter provides an overview of the state of the art of the intersections between three multidisciplinary disciplines. First, a de nition of Design Research is provided, together with its main trends and open issues. Then, the review focuses on the contribution of Design Research to the HRI eld, which can be summed up in actions focused on three aspects: artefacts, stakeholders, and contexts. This is followed by a focus on the role of Design Research within the context of children studies, in which it is possible to identify two main design-child relationships: design as a method for developing children’s learning experiences; and children as part of the design process for developing novel interactive systems. The third chapter introduces the Research through Design (RtD) approach and its relevance in conducting design research in HRI. The proposed methodology, based on this approach, is particularly characterized by the presence of design explorations as study methods. These, in turn, are developed through a common project’s methodology, also reported in this chapter. The fourth chapter is dedicated to the analysis of the scenario in which the child-robot interaction takes place. This was aimed at understanding what is edutainment robotics for children, its common features, how it relates to existing children play types, and where the interaction takes place. The chapter provides also a focus on the relationship between children and technology on a more general level, through which two themes and relative design opportunities were identi ed: physically active play and objects-to-think-with. These were respectively addressed in the two design explorations presented in this thesis: Phygital Play and Shybo. The Phygital Play project consists of an exploration of natural interaction modalities with robots, through mixed-reality, for fostering children’s active behaviours. To this end, a game platform was developed for allowing children to play with or against a robot, through body movement. Shybo, instead, is a low-anthropomorphic robot for playful learning activities with children that can be carried out in educational contexts. The robot, which reacts to properties of the physical environment, is designed to support different kinds of experiences. Then, the chapter eight is dedicated to the research outcomes, that were de ned through a process of re ection. The contribution of the research was analysed and documented by focusing on three main levels, namely: artefact, knowledge and theory. The artefact level corresponds to the situated implementations developed through the projects. The knowledge level consists of a set of actionable principles, emerged from the results and lessons learned from the projects. At the theory level, a theoretical framework was proposed with the aim of informing the future design of child- robot play applications. The last chapter provides a nal overview of the doctoral research, a series of limitations regarding the research, its process and its outcomes, and some indications for future research.
  • Preprint
    Understanding passenger intents and extracting relevant slots are important building blocks towards developing contextual dialogue systems for natural interactions in autonomous vehicles (AV). In this work, we explored AMIE (Automated-vehicle Multi-modal In-cabin Experience), the in-cabin agent responsible for handling certain passenger-vehicle interactions. When the passengers give instructions to AMIE, the agent should parse such commands properly and trigger the appropriate functionality of the AV system. In our current explorations, we focused on AMIE scenarios describing usages around setting or changing the destination and route, updating driving behavior or speed, finishing the trip and other use-cases to support various natural commands. We collected a multi-modal in-cabin dataset with multi-turn dialogues between the passengers and AMIE using a Wizard-of-Oz scheme via a realistic scavenger hunt game activity. After exploring various recent Recurrent Neural Networks (RNN) based techniques, we introduced our own hierarchical joint models to recognize passenger intents along with relevant slots associated with the action to be performed in AV scenarios. Our experimental results outperformed certain competitive baselines and achieved overall F1 scores of 0.91 for utterance-level intent detection and 0.96 for slot filling tasks. In addition, we conducted initial speech-to-text explorations by comparing intent/slot models trained and tested on human transcriptions versus noisy Automatic Speech Recognition (ASR) outputs. Finally, we compared the results with single passenger rides versus the rides with multiple passengers.
  • Article
    Robotic devices for older adults are becoming a reality. New robots are being introduced for the growing subpopulation of healthy older adults, with an emphasis on supporting the positive aspects of aging. In order to inform the design and implementation of such robots, the relevant needs and concerns of this population should be studied, mapped, and translated into recommendations. We present a qualitative study of thirty cognitively-intact older adults, evaluating their attitudes and emotional reactions towards different types of home robotic devices. Interview analysis of participants reactions to videos of six devices uncovered four user needs that can be threatened by the introduction of home robots: the need for independence, the need for control, the fear of being replaced, and the need for authenticity. Furthermore, results reveal that cognitively-intact older adults are willing to adopt robotic devices into their homes, contingent upon their preferences and concerns being addressed. We provide recommendations regarding how researchers and designers of home robots can better address the user needs of healthy older adults by leveraging aspects of the robot's function, speech, appearance, size, proactivity, and mobility.
  • Article
    Full-text available
    Driver steering performance is strongly related to gaze behavior in natural driving. The investigation of gaze behavior is crucial for understanding driver response to driver-automation shared control, which was developed to reduce the workload of drivers while keeping them in the control loop. This study focused on driver gaze behavior and its relationship to steering performance when driving with different levels of shared control. Shared control was achieved through a haptic guidance steering system, which provided active assistant torques on the steering wheel. A driving simulator experiment was conducted with 15 participants, who drove five trials, each with a different level of shared control: (1) no haptic guidance, (2) normal haptic guidance, (3) strong haptic guidance, (4) having a driver's hands hovering over the steering wheel as a standby during automated driving, and (5) hands free during automated driving. The results show a tendency that the correlation between gaze and steering movements decreased when the automation authority of shared control increased. A reduction in the lead time of gaze over steering movement was found in the condition of strong haptic guidance compared to normal haptic guidance and manual driving. This study suggests the potential of using the relationship between gaze behavior and steering performance to analyze driver interaction with automation.
  • Conference Paper
    Full-text available
    The comfort level of passengers is an important factor in measuring user experience in any form of transportation, including in autonomous vehicles. One of the main factors that determines user acceptance of autonomous vehicles is the passenger's level of discomfort in a 'control-and authority-less' experience. In this paper, we propose an approach for formulating discomfort through 'on-the-road' field studies, with human driven vehicles, while the passenger provides real-time explicit feedback on discomfort via a potentiometer. While previous studies focused on the association between vehicle dynamics and passenger discomfort, we demonstrate here how we can improve the classification ability of passenger discomfort by employing a multi-dimensional model that also takes into account the external scenario (contextual information). This is achieved by processing image data (e.g. distance from nearest bicycle) recorded through an outward looking camera in addition to location/route data obtained from other sensors like GPS. As such, the focus of this paper is on classification of external information.
  • Conference Paper
    We describe the development and user testing of an ambient display for autonomous vehicles. Instead of providing feedback about driving actions, once executed, it communicates driving decisions in advance, via light signals in passengers" peripheral vision. This ambient display was tested in an WoZ-based on-the-road-driving simulation of a fully autonomous vehicle. Findings from a preliminary study with 14 participants suggest that such a display might be particularly useful to communicate upcoming inertia changes for passengers.
  • Conference Paper
    Full-text available
    Simulator studies have been conducted in the automotive domain since the 1960s. Recently, automated driving studies have become more popular as real-world automated cars start to emerge but at this time not all levels of automation can be realized. A simulation does not entail all details of real driving, creating a realistic simulation experience - both on a psychological and physical level - proposes recurring challenges. These are among others: sample acquisition, simulator sickness, simulator training, interface design, take-over requests and secondary tasks in automated driving simulator studies. In this paper, we review existing literature and summarize important lessons from simulations in the domain of driving automation to provide considerations for studies investigating driver behavior in the age of highly automated driving.
  • Conference Paper
    Researchers and designers of in-vehicle interactions and interfaces currently have to choose between performing evaluation and human factors experiments in laboratory driving simulators or on-road experiments. To enjoy the benefit of customizable course design in controlled experiments with the immediacy and rich sensations of on-road driving, we have developed a new method and tools to enable VR driving simulation in a vehicle as it travels on a road. In this paper, we describe how the cost-effective and flexible implementation of this platform allows for rapid prototyping. A preliminary pilot test (N = 6), centered on an autonomous driving scenario, yields promising results, illustrating proof of concept and indicating that a basic implementation of the system can invoke genuine responses from test participants.
  • Article
    Full-text available
    Trust is important for any relationship, especially so with self-driving vehicles: passengers must trust these vehicles with their life. Given the criticality of maintaining passenger’s trust, yet the dearth of self-driving trust repair research relative to the growth of the self-driving industry, we conducted two studies to better understand how people view errors committed by self-driving cars, as well as what types of trust repair efforts may be viable for use by self-driving cars. Experiment 1 manipulated error type and driver types to determine whether driver type (human versus self-driving) affected how participants assessed errors. Results indicate that errors committed by both driver types are not assessed differently. Given the similarity, experiment 2 focused on self-driving cars, using a wide variety of trust repair efforts to confirm human-human research and determine which repairs were most effective at mitigating the effect of violations on trust. We confirmed the pattern of trust repairs in human-human research, and found that some apologies were more effective at repairing trust than some denials. These findings help focus future research, while providing broad guidance as to potential methods for approaching trust repair with self-driving cars.
  • Conference Paper
    Full-text available
    Autonomous vehicles have been in development for nearly thirty years and recently have begun to operate in real-world, uncontrolled settings. With such advances, more widespread research and evaluation of human interaction with autonomous vehicles (AV) is necessary. Here, we present an interview study of 32 pedestrians who have interacted with Uber AVs. Our findings are focused on understanding and trust of AVs, perceptions of AVs and artificial intelligence, and how the perception of a brand affects these constructs. We found an inherent relationship between favorable perceptions of technology and feelings of trust toward AVs. Trust in AVs was also influenced by a favorable interpretation of the company's brand and facilitated by knowledge about what AV technology is and how it might fit into everyday life. To our knowledge, this paper is the first to surface AV-related interview data from pedestrians in a natural, real-world setting.
  • Article
    The phrase "in-the-wild" is becoming popular again in the field of human-computer interaction (HCI), describing approaches to HCI research and accounts of user experience phenomena that differ from those derived from other lab-based methods. The phrase first came to the forefront 20-25 years ago when anthropologists Jean Lave (1988), Lucy Suchman (1987), and Ed Hutchins (1995) began writing about cognition being in-the-wild. Today, it is used more broadly to refer to research that seeks to understand new technology interventions in everyday living. A reason for its resurgence in contemporary HCI is an acknowledgment that so much technology is now embedded and used in our everyday lives. Researchers have begun following suit—decamping from their usability and living labs and moving into the wild; carrying out in-situ development and engagement, sampling experiences, and probing people in their homes and on the streets. The aim of this book is to examine what this new direction entails and what it ...
  • Autonomous vehicle arrangement and method for controlling an autonomous vehicle
    • Ralf Bergholz
    • Klaus Timm
    • Hubert Weisser
    Ralf Bergholz, Klaus Timm, and Hubert Weisser. 2000. Autonomous vehicle arrangement and method for controlling an autonomous vehicle. (Nov. 21 2000). http://www.google.com/patents/US6151539 US Patent 6,151,539.
  • Springer Handbook of Automation Chapter Human Factors in Automation Design
    • D John
    • Bobbie D Lee
    • Seppelt
    John D. Lee and Bobbie D. Seppelt. 2009. Springer Handbook of Automation. Springer Berlin Heidelberg, Berlin, Heidelberg, Chapter Human Factors in Automation Design, 417–436.
  • Autonomous Vehicle Implementation Predictions
    • Todd Litman
    Todd Litman. 2014. Autonomous Vehicle Implementation Predictions. Victoria Transport Policy Institute 28 (2014). http://sh.st/st/ 787f28ed3e745c14417e4aec27303038/http: //www.vtpi.org/avip.pdf
  • Somebody Grab the Wheel: State Autonomous Vehicle Legislation and the Road to a National Regime
    • R Andrew
    • Swanson
    Andrew R Swanson. 2013. Somebody Grab the Wheel: State Autonomous Vehicle Legislation and the Road to a National Regime. Marq. L. Rev. 97 (2013), 1085.
  • Tire and Vehicle Dynamics. Butterworth-Heinemann
    • B Hans
    • Pacejka
    • Michael Montemerlo
    • Jan Becker
    • Suhrid Bhat
    • Hendrik Dahlkamp
    • Dmitri Dolgov
    • Scott Ettinger
    • Dirk Haehnel
    • Tim Hilden
    • Gabe Hoffmann
    • Burkhard Huhnke
    • Doug Johnston
    • Stefan Klumpp
    • Dirk Langer
    • Anthony Levandowski
    • Jesse Levinson
    • Julien Marcil
  • Conference Paper
    Full-text available
    In future automated driving systems, drivers will be free to perform other secondary tasks, not needing to stay vigilant in monitoring the car’s activity. However, there will still be situations in which drivers are required to take-over control of the vehicle, most likely from a highly distracted state. While highly automated vehicles would ideally accommodate structured takeovers, providing ample time and warning, it is still very important to examine how drivers would behave when they are subjected to an unstructured emergency transition of control. In this study, we observed how participants (N=30) in a driving simulator performed after they experienced a loss of automation. We tested three transition time conditions, with an unstructured transition of control occurring 2 seconds, 5 seconds, or 8 seconds before the participants encountered a road hazard that required the drivers’ intervention. Participants were given a passive distraction (watching a video) to do while the automated driving mode was enabled, so they needed to disengage from the task and regain control of the car when the transition occurred. Few drivers in the 2 second condition were able to safely negotiate the road hazard situation, while the majority of drivers in the 5 or 8 second conditions were able to navigate the hazard safely. Similarly, drivers in the 2 second condition rated the vehicle to be less trustworthy than drivers in the 5 and 8 second conditions. From the study results, we are able to narrow down a minimum amount of time in which drivers can take over the control of vehicle safely and comfortably from the automated system in the advent of an impending road hazard.
  • Conference Paper
    Full-text available
    The immense technological advancements in the past decade have enabled robots to enjoy high levels of autonomy, paving their way into our society. The recent catastrophic accidents involving autonomous systems (e.g., Tesla fatal car accident), however, show that sole engineering progress in the technology is not enough to guarantee a safe and productive partnership between a human and a robot. In this paper we argue that we also need to advance our understanding of the role of social trust within human-robot relationships, and formulate a theory for expressing and reasoning about trust in the context of decisions affecting collaboration or competition between humans and robots. Therefore, we call for cross-disciplinary collaborations to study the formalization of social trust in the context of human-robot relationship. We lay the groundwork for such a study in this paper.
  • Article
    Car automation promises to free our hands from the steering wheel but might demand more from our minds.
  • Conference Paper
    Full-text available
    This video introduces a methodology for simulating an autonomous vehicle on open public roads. The video showcases participant reaction footage collected in the RRADS (Real Road Autonomous Driving Simulator). Although our study using this simulator did not use overt deception--the consent form clearly states that a licensed driver is operating the vehicle--the protocol was designed to support suspension of disbelief. Several participants who did not read the consent form clearly strongly believed that the vehicle was autonomous; this provides a lens onto the attitudes and concerns that people in real-world autonomous vehicles might have, and also points to ways that a protocol that deliberately used misdirection could gain ecologically valid reactions from study participants.
  • Conference Paper
    How will pedestrians and cyclists interact with self-driving cars when there is no human driver? To find answers to this question we need a secure experimental design in which pedestrians can interact with a car that appears to drive on its own. In Ghost Driver we staged a fake autonomous car by installing LIDARs, cameras and decals on the outside of the vehicle and by covering the driver with a seat costume so that it appeared that there was no driver in the car. In initial field studies we found that this Wizard-of-Oz technique convinced more than 80% of the participants that the car was driving autonomously without a driver. Consequently the Ghost Driver methodology could become a platform for further investigation of how pedestrians or cyclists interact with driverless vehicles.
  • Book
    For the past hundred years, innovation within the automotive sector has created safer, cleaner, and more affordable vehicles, but progress has been incremental. The industry now appears close to substantial change, engendered by autonomous, or "self-driving," vehicle technologies. This technology offers the possibility of significant benefits to social welfare — saving lives; reducing crashes, congestion, fuel consumption, and pollution; increasing mobility for the disabled; and ultimately improving land use. This report is intended as a guide for state and federal policymakers on the many issues that this technology raises. After surveying the advantages and disadvantages of the technology, RAND researchers determined that the benefits of the technology likely outweigh the disadvantages. However, many of the benefits will accrue to parties other than the technology's purchasers. These positive externalities may justify some form of subsidy. The report also explores policy issues, communications, regulation and standards, and liability issues raised by the technology; and concludes with some tentative guidance for policymakers, guided largely by the principle that the technology should be allowed and perhaps encouraged when it is superior to an average human driver.
  • Conference Paper
    Full-text available
    This platform paper introduces a methodology for simulating an autonomous vehicle on open public roads. The paper outlines the technology and protocol needed for running these simulations, and describes an instance where the Real Road Autonomous Driving Simulator (RRADS) was used to evaluate 3 prototypes in a between-participant study design. 35 participants were interviewed at length before and after entering the RRADS. Although our study did not use overt deception---the consent form clearly states that a licensed driver is operating the vehicle---the protocol was designed to support suspension of disbelief. Several participants who did not read the consent form clearly strongly believed that they were interacting with a fully autonomous vehicle. The RRADS platform provides a lens onto the attitudes and concerns that people in real-world autonomous vehicles might have, and also points to ways that a protocol deliberately using misdirection can gain ecologically valid reactions from study participants.
  • Article
    In this new paperback edition of Tyre and Vehicle Dynamics, theory is supported by practical and experimental evidence. Pacejka provides both basic and advanced explanations of the pneumatic tyre and its impact on vehicle dynamic performance. The book shows the way in which tyre models are incorporated in vehicle models and how important tyre influence is on overall vehicle behaviour. Those working in any industry involving equipment with tyres will continue to find this book both extremely relevant and useful. * Written by a world expert in tyre dynamics * Covers both basic and advanced tyre modelling and simulation, including case studies of application examples and chapter exercises * Indispensable for any engineer working in vehicle system dynamics and for any industry involving equipment with tyres.
  • Article
    Up to now, the resulting restrictions for HMI development from the limited availability of technical components lead to a method where a technological setup is avoided: In the Wizard of Oz Technique a human being takes over the role of the technical system components and simulates complex processes, e.g. the recognition and interpretation of speech inputs. In this field, this method was widespread in the 1990s, when computing power limited speech processing. Wizard of Oz experiments provide a possibility for user testing of future technologies which are not yet available in hardware. For automotive issues, first adaptations into virtual reality environments and driving simulators already exist. This paper describes the transfer of this experimental technique into a driving vehicle: For research and development of driver assistance systems a Wizard of Oz vehicle was developed. In this vehicle, a hidden second driver in the load compartment is able to take over parts of the driving task varying the degree of automation from manual control to complete automatic control. Besides alternative control of the driving subtasks a cooperative control of the vehicle is possible by simultaneous manipulation of driver and wizard. This setup allows the person on the driver's seat to experience not yet existing future driver assistance systems. Hence, sensor development issues are no longer limiting user testing and different degrees of " intelligence" of the driver assistance system can be realized before spending effort in technical development. Results from user testing can be generated much earlier in the development process and find their entry into the ongoing system design. Besides the advantages for early usability testing, this instrument offers outstanding possibilities in the development of assistance functionalities as even very early ideas can be experienced in a real car. Therefore, the Wizard of Oz vehicle is an important instrument within a rapid prototyping approach in the development of assistance systems, completing established development tools such as driving simulators and prototype vehicles. By means of the description of the technical realization of the experimental vehicle the conception and design of user studies is presented. Here, two example studies designed for validation of the tool are introduced and described: First, an in-vehicle information system was simulated. The system evaluation of 18 users was compared to the evaluation of a real system setup. In the second experiment a lanekeeping system is emulated as an example for intervening driver assistance. 19 users evaluated both the Wizard of Oz system and a prototype lanekeeping system. The results provide the background for a discussion of the advantages and constraints of using the Wizard of Oz-technique in a vehicle for designing HMIs of future driver assistance.
  • Technical Report
    Full-text available
    The Concepts of Operation document evaluates the functional framework of operations for Level 2 and Level 3 automated vehicle systems. This is done by defining the varying levels of automation, the operator vehicle interactions, and system components; and further, by assessing the automation relevant parameters from a scenario-based analysis stand-point. Specific to the “Human Factors Evaluation of Level 2 and Level 3 Automated Driving Concepts” research effort, scenarios and literature are used to identify the range of near- to mid-term production-intent systems such that follow-on research topics with highest impact potential can be identified through commonalities in operational concepts.
  • Article
    Raising the automation level in cars is an imaginable scenario for the future in order to improve traffic safety. However, as long as there are situations that cannot be handled by the automation, the driver has to be enabled to take over the driving task in a safe manner. The focus of the current study is to understand at which point in time a driver’s attention must be directed back to the driving task. To investigate this issue, an experiment was conducted in a dynamic driving simulator and two take-over times were examined and compared to manual driving. The conditions of the experiment were designed to examine the take-over process of inattentive drivers engaged in an interaction with a tablet computer. The results show distinct automation effects in both take-over conditions. With shorter take-over time, decision making and reactions are faster but generally worse in quality.
  • 2015. I rode 500 miles in a self-driving car and saw the future. It's delightfully dull
    • Alexander Davies
    Alexander Davies. 2015. I rode 500 miles in a self-driving car and saw the future. It's delightfully dull. (2015).
  • Article
    Many researchers use Wizard of Oz (WoZ) as an experimental technique, but there are methodologi-cal concerns over its use, and no comprehensive criteria on how to best employ it. We systematically review 54 WoZ experiments published in the primary HRI publication venues from 2001 -2011. Us-ing criteria proposed by Fraser and Gilbert (1991), Green et al. (2004), Steinfeld et al. (2009), and Kelley (1984), we analyzed how researchers conducted HRI WoZ experiments. Researchers mainly used WoZ for verbal (72.2%) and non-verbal (48.1%) processing. Most constrained wizard produc-tion (90.7%), but few constrained wizard recognition (11%). Few reported measuring wizard error (3.7%), and few reported pre-experiment wizard training (5.4%). Few reported using WoZ in an iterative manner (24.1%). Based on these results we propose new reporting guidelines to aid future research.
  • Article
    Racecar drivers have the ability to operate a vehicle at its friction limit without losing control. If autonomous vehicles or driver assistance systems had similar capabilities, many fatal accidents could be avoided. To advance this goal, an autonomous racing controller is designed to gain insights into vehicle control at the friction limits. A bicycle model and a 'g-g' diagram are used to mimic racecar drivers' internal vehicle model. Lanekeeping steering feedback and wheel slip feedback controllers are used to imitate drivers making steering and throttle corrections according to the vehicle responses. Experimental results on a low friction surface demonstrate that the controller can robustly track a path while operating at the limits of tyre adhesion and provide insights for the future development of vehicle safety systems.
  • Conference Paper
    We discuss current approaches to the development of natural language dialogue systems, and claim that they do not sufficiently consider the unique qualities of man-machine interaction as distinct from general human discourse. We conclude that empirical studies of this unique communication situation is required for the de-velopment of user-friendly interactive systems. One way of achieving this is through the use of so-called Wizard of Oz studies. We describe our work in this ar-ea. The focus is on the practical execution of the studies and the methodological conclusions that we have drawn on the basis of our experience. While the focus is on natural language interfaces, the methods used and the conclusions drawn from the results obtained are of relevance also to other kinds of intelligent interfaces. 1 THE NEED FOR WIZARD OF OZ STUDIES Dialogue has been an active research area for quite some time in natural language processing. It is fair to say that researchers studying dialogue and discourse have developed their theories through detailed analysis of empirical data from many di-
  • Article
    Full-text available
    In the recent years, intelligent vehicles became an important part of the service robotics. Besides routine tasks for the service robotics, such as localization and navigation, intelligent vehicle must solve many specific tasks. Although fully autonomous intelligent vehicle does not exist, there are many support systems nowadays for which vehicles are closer to the vision of self-piloted vehicle. This article presents state of art in this section of service robotics. The main body of the article introduces a definition of intelligent vehicle and automated functions used in such vehicles. Future challenges of the intelligent vehicles are also presented.
  • Article
    Full-text available
    The validity of driving simulators for behavioral research is discussed. The concept of validity is introduced and explained, and a survey of validation studies follows, in the TNO driving simulator and others, comparing field and simulator study results. Results for mid-level driving simulators show that generally absolute validity of route choice behavior is obtained and relative validity of speed and lateral control behavior is obtained. There is evidence suggesting that for a number of applications the presence of a moving base and possibly a higher image resolution might increase the validity of a driving simulator.
  • Article
    The validity of research results obtained using the fixed-base vehicle simulator of the Institute for Perception TNO was studied during straight-road driving. Absolute and relative validities were mainly evaluated in terms of system performance and driver behavior for inexperienced and experienced drivers, who had to perform lateral and longitudinal vehicle control both in the simulator and in an instrumented car on the road. Task demands for each control were varied with a free and forced accuracy instruction. Overall results showed good absolute and relative validity for longitudinal vehicle control; lateral vehicle control offered good relative validity. Lateral control performance lacked absolute validity due to the drivers' diminished perception of lateral translations (absence of kinesthetic feedback). Drivers were easily able to perceive yaw rotations in the simulator. Performance in the simulator was a more sensitive discriminator of driving experience than was performance in the instrumented car on the road.
  • Article
    Driving performance in an instrumented vehicle was compared with performance in a low-cost, fixed-based driving simulator. Six men and six women drove a freeway route while periodically dialling simulated phone calls. The same subjects drove a laboratory driving simulator using two visual fidelity levels: a colour scene with relatively high detail, and a monochrome (night) scene showing only road-edge markings. Lane position, speed, steering-wheel angle and throttle position were recorded in both contexts. Lane-keeping in the simulator was less precise than on the road, but speed control performance was comparable. The SD of lane position in normal driving was about twice as large, on average, in the simulator (0.360 versus 0.165 m). Lane keeping and speed control were less precise when dialling the phone than in normal driving, both in the simulator and on the road, but the performance decrement was greater in the simulator. The addition of the phone task increased the mean lateral speed in the car by about 43%, while in the simulator the mean lateral speed increased by 158% with the addition of the phone task. Subjects >60 years of age showed larger performance decrements during a concurrent phone dialling task than did subjects 20–30 years of age both in the simulator and on-road. No important differences in driving performance were found between the high and low simulator scene fidelity levels. The simulator demonstrated good absolute validity for measures of speed control and good relative validity for the effects of the phone task and age on driving precision.
  • Article
    Boss is an autonomous vehicle that uses on-board sensors (global positioning system, lasers, radars, and cameras) to track other vehicles, detect static obstacles, and localize itself relative to a road model. A three-layer planning system combines mission, behav-ioral, and motion planning to drive in urban environments. The mission planning layer considers which street to take to achieve a mission goal. The behavioral layer determines when to change lanes and precedence at intersections and performs error recovery maneu-vers. The motion planning layer selects actions to avoid obstacles while making progress toward local goals. The system was developed from the ground up to address the require-ments of the DARPA Urban Challenge using a spiral system development process with a heavy emphasis on regular, regressive system testing. During the National Qualifica-tion Event and the 85-km Urban Challenge Final Event, Boss demonstrated some of its capabilities, qualifying first and winning the challenge. C 2008 Wiley Periodicals, Inc.
  • Article
    Firmata is a generic protocol for communicating with mi-crocontrollers from software on a host computer. The cen-tral goal is to make the microcontroller an extension of the programming environment on the host computer in a man-ner that feels natural in that programming environment. It was designed to be open and flexible so that any program-ming environment can support it, and simple to implement both on the microcontroller and the host computer to ensure a wide range of implementations. The current reference im-plementation is a library for Arduino/Wiring and is included with Arduino software package since version 0012. There are matching software modules for a number of languages, like Pd, OpenFrameworks, Max/MSP, and Processing.
  • Article
    This article presents the architecture of Junior, a robotic vehicle capable of navigating urban environments autonomously. In doing so, the vehicle is able to select its own routes, perceive and interact with other traffic, and execute various urban driving skills including lane changes, U-turns, parking, and merging into moving traffic. The vehicle successfully finished and won second place in the DARPA Urban Challenge, a robot competition organized by the U.S. Government. © 2008 Wiley Periodicals, Inc.
  • Conference Paper
    Full-text available
    In this paper, the theater-system technique, a method for agile designing and testing of system behavior and interaction concepts is described. The technique is based on the Wizard-of-Oz approach, originally used for emulating automated speech recognition, and is extended towards an interactive, user-centered design technique. The paper describes the design process using the theater-system technique, the technical build-up of the theater-system, and an application of the technique: the design of a haptic-multimodal interaction strategy for highly automated vehicles. The use of the theater-system in the design process is manifold: It is used for the concrete design work of the design team, for the assessment of user expectations as well as for early usability assessments, extending the principles of user-centered design towards a dynamically balanced design.
  • Conference Paper
    Full-text available
    This paper presents a nonlinear control law for an automobile to autonomously track a trajectory, provided in real-time, on rapidly varying, off-road terrain. Existing methods can suffer from a lack of global stability, a lack of tracking accuracy, or a dependence on smooth road surfaces, any one of which could lead to the loss of the vehicle in autonomous off-road driving. This work treats automobile trajectory tracking in a new manner, by considering the orientation of the front wheels - not the vehicle's body - with respect to the desired trajectory, enabling collocated control of the system. A steering control law is designed using the kinematic equations of motion, for which global asymptotic stability is proven. This control law is then augmented to handle the dynamics of pneumatic tires and of the servo-actuated steering wheel. To control vehicle speed, the brake and throttle are actuated by a switching proportional integral (PI) controller. The complete control system consumes a negligible fraction of a computer's resources. It was implemented on a Volkswagen Touareg, "Stanley", the Stanford Racing Team's entry in the DARPA Grand Challenge 2005, a 132 mi autonomous off-road race. Experimental results from Stanley demonstrate the ability of the controller to track trajectories between obstacles, over steep and wavy terrain, through deep mud puddles, and along cliff edges, with a typical root mean square (RMS) crosstrack error of under 0.1 m. In the DARPA National Qualification Event 2005, Stanley was the only vehicle out of 40 competitors to not hit an obstacle or miss a gate, and in the DARPA Grand Challenge 2005 Stanley had the fastest course completion time.
  • Conference Paper
    Full-text available
    User experience evaluation in human-robot interaction is most often an expensive and difficult task. To allow the evaluation of various factors and aspects of user experience, a fully functional (humanoid) robot is recommended. This work presents technical and methodological considerations on the applicability of the Wizard of Oz (WOz) approach to enable user experience evaluation in the field of Human-Robot Interaction. We briefly describe the technical aspects of the setup, the applicability of the method, and a first case study using this methodological approach to gain an early understanding of the user experience factors that are important for the development of a human-humanoid interaction scenario.
  • Article
    A robotic platform for mapping of weed populations in fields was used to demonstrate intelligent concepts for autonomous vehicles in agriculture which may eventually result in a new sustainable model for developed agriculture. The vehicle presented here is adapted to operate in 0·25 and row crops and equipped with cameras for row guidance and weed detection. A modular approach is taken with four identical wheel modules, allowing four wheel steering and propulsion of the platform. The result is improved mobility which allows parallel displacement of the vehicle during turns by decoupling adjustments in position from adjustments in orientation. Control of the platform is provided through a vehicle electronics and control system based on embedded controllers and standard communication protocols. The software implements a hybrid deliberate software architecture that allows a hierarchical decomposition of the operation. The lowest level implements a reactive feedback control mechanism based on an extension of simple control for car-like vehicles to the four wheel case. The controller design forces the front and rear of the vehicle to follow a pre-determined path and allows the vehicle to maintain a fixed orientation relative to the path. The controller rationale is outlined and results from experiments in the field are presented.
  • Conference Paper
    We are interested in exploring how robots controlled using Wizard of Oz (WoO) should interrupt humans in various social settings. While there is considerable work on interruption and interruptibility in HCI, little has been done to explore how these concepts will map robotic interaction. As part of our efforts to investigate interruption and interruptibility in HRI we used WoO-based methodology to investigate robot behaviours in a simple interruption scenario. In this report we contribute a design critique that discusses this methodology, and common concerns that could be generalized to other social HRI experiments as well as reflections on our future interruption HRI research.
  • Article
    Data from on-road and simulation studies were compared to assess the validity of measures generated in the simulator. In the on-road study, driver interaction with three manual address entry methods (keypad, touch screen and rotational controller) was assessed in an instrumented vehicle to evaluate relative usability and safety implications. A separate group of participants drove a similar protocol in a medium fidelity, fixed-base driving simulator to assess the extent to which simulator measures mirrored those obtained in the field. Visual attention and task measures mapped very closely between the two environments. In general, however, driving performance measures did not differentiate among devices at the level of demand employed in this study. The findings obtained for visual attention and task engagement suggest that medium fidelity simulation provides a safe and effective means to evaluate the effects of in-vehicle information systems (IVIS) designs on these categories of driver behaviour. STATEMENT OF RELEVANCE: Realistic evaluation of the user interface of IVIS has significant implications for both user acceptance and safety. This study addresses the validity of driving simulation for accurately modelling differences between interface methodologies by comparing results from the field with those from a medium fidelity, fixed-base simulator.
  • Technical Report
    Full-text available
    Good design is not free of form. It does not necessarily happen through a mere sampling of technologies packaged together, through pure analysis, or just by following procedures. Good design begins with inspiration and a vision, a mental image of the end product, which can sometimes be described with a design metaphor. A successful example from the 20th century is the desktop metaphor, which took a real desktop as an orientation for the manipulation of electronic documents on a computer. Initially defined by Xerox, then refined by Apple and others, it could be found on almost every computer by the turn of the 20th century. This paper sketches a specific metaphor for the emerging field of highly automated vehicles, their interactions with human users and with other vehicles. In the introduction, general questions on vehicle automation are raised and related to the physical control of conventional vehicles and to the automation of some late 20th century vehicles. After some words on design metaphors, the H-Metaphor is introduced. More details of the metaphor's source are described and their application to human-machine interaction, automation and management of intelligent vehicles sketched. Finally, risks and opportunities to apply the metaphor to technical applications are discussed.
  • Article
    Adults aged 67 to 78 were compared in driving performance on the road and in driving simulators. The driving simulator operated in an urban environment and required the subjects to execute maneuvers that would allow evaluation of their driving ability in a multitask situation. The two tests proved to be strongly and negatively correlated - the higher the score on the road test, the lower the score on the driving simulator.
  • Article
    The behavioral validation of an advanced driving simulator for its use in evaluating speeding countermeasures was performed for mean speed. Using mature drivers, 24 participants drove an instrumented car and 20 participants drove the simulator in two separate experiments. Participants drove on roads which contained transverse rumble strips at three sites, as well as three equivalent control sites. The three pairs of sites involved deceleration, and were the approaches to stop sign intersections, right curves, and left curves. Numerical correspondence (absolute validity), relative correspondence (or validity), and interactive (or dynamic) relative validity were analyzed, the latter using correlations developed from canonical correlation. Participants reacted to the rumble strips, in relation to their deceleration pattern on the control road, in very similar ways in both the instrumented car and simulator experiments, establishing the relative validities. However, participants generally drove faster in the instrumented car than the simulator, resulting in absolute validity not being established.
  • Conference Paper
    Full-text available
    Nonverbal communication plays an important role in coordinating teammates' actions for collaborative activities. In this paper, we explore the impact of non-verbal social cues and behavior on task performance by a human-robot team. We report our results from an experiment where naive human subjects guide a robot to perform a physical task using speech and gesture. Both self-report via questionnaire and behavioral analysis of video offer evidence to support our hypothesis that implicit non-verbal communication positively impacts human-robot task performance with respect to understandability of the robot, efficiency of task performance, and robustness to errors that arise from miscommunication.