Toshiaki Kakinami

3D parking assistant system

We present the evolution of our prototype for a new generation of circumstance recognition system. This prototype provides the driver a 3D representation of the scene surrounding his vehicle. We focus essentially on the improvement of quality and robustness of the set of 3D points reconstructed by tracking from acquisitions and on a process of obstacle modeling from this set of 3D points to achieve a good degree of realism.

Ontology-based decision making on uncontrolled intersections and narrow roads

Many Advanced Driver Assistance Systems (ADAS) have been developed to improve car safety. However, it is still a challenging problem to make autonomous vehicles to drive safely on urban streets such as uncontrolled intersections (without traffic lights) and narrow roads. In this paper, we introduce a decision making system that can assist autonomous vehicles at uncontrolled intersections and narrow roads. We constructed a machine understandable ontology-based Knowledge Base, which contains maps and traffic regulations. The system makes decisions in comply with traffic regulations such as Right-Of-Way rules when it receives a collision warning signal. The decisions are sent to a path planning system to change the route or stop to avoid collisions.

AUTONOMOUS VEHICLE CONTROL SYSTEM USING AN IMAGE PROCESSING SENSOR

This paper presents an inter-vehicle distance control system which controls the vehicles throttle and braking functions. It does this by detecting the position of the driving lane and the distance to the vehicle in front based on the images obtained by a CCD camera.

Movement Detection for Safer Backward Maneuvers

Backing-up accidents are responsible for about 120 deaths and more than 6000 injuries every year in the US. IMRA-Europe has been working on a static obstacles detection system for a few years now. We propose to complement this with a system capable of detecting and localizing in real-time moving obstacles even when the vehicle is moving. This system is based on camera movement compensation using IPM (inverse perspective mapping) and probabilistic reasoning for movement detection. Using shape filtering our approach can even detect only specific types of moving objects (pedestrians, cars...). In this paper, we present the algorithms developed, show some results of the system and conclude on which future developments will bring performance improvements