• [Show abstract] [Hide abstract]
    ABSTRACT: The future mobility of urban areas is changing constantly; ideally, vehicles should be able to drive autonomously through traffic. Unfortunately, autonomous vehicles are not yet fully capable of matching human performance. Therefore, the teleoperation of vehicles presents a solution for this task. During teleoperation, a human driver is responsible for driving the vehicle using information transmitted from the vehicle to a working station. Unfortunately, because of the artificial environment in which the operator is located, it is very difficult to achieve high telepresence and accurate speed estimation. It is known that in order to safely drive a vehicle, it is very important to be able to correctly estimate the vehicle's speed. This paper presents a study conducted to quantify the speed perception tendency of a human operator at the working station. Additionally, it is shown that a training process can at least temporarily improve speed perception. Furthermore, the implementation of zoom blur to increase optical flow is shown to positively influence speed perception. Four hypotheses are defined and analysed to study speed perception at an operator's working station. The results are presented and discussed.
    International Journal of Advanced Robotic Systems 01/2013; 10. DOI:10.5772/56735 · 0.50 Impact Factor
  • Source
    Massachusetts Institute of Technology, 07/2011, Degree: PhD
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This work tackles the risk estimation problem from a new perspective: a framework is proposed for reasoning about traffic situations and collision risk at a semantic level, while classic approaches typically reason at a trajectory level. Risk is assessed by estimating the intentions of drivers and detecting conflicts between them, rather than by predicting the future trajectories of the vehicles and detecting collisions between them. More specifically, dangerous situations are identified by comparing what drivers intend to do with what they are expected to do according to the traffic rules. The reasoning about intentions and expectations is performed in a probabilistic manner, in order to take into account sensor uncertainties and interpretation ambiguities. This framework can in theory be applied to any type of traffic situation; here we present its application to the specific case of road intersections. The proposed motion model takes into account the mutual influences between the maneuvers performed by vehicles at an intersection. It also incorporates information about the influence of the geometry and topology of the intersection on the behavior of a vehicle, and therefore can be applied to arbitrary intersection layouts. The approach was validated with field trials using passenger vehicles equipped with Vehicle-to-Vehicle wireless communication modems, and in simulation. The results demonstrate that the algorithm is able to detect dangerous situations early and complies with real-time constraints.