Integrating behavioral, perceptual, and world knowledge in reactive navigation
Robotics and Autonomous Systems
| Hi-index | 0.98 |
An architecture to plan and control the motion of a car-like vehicle is presented. With this architecture, the vehicle is able to avoid stationary and moving obstacles in its local environment while obeying the highway code and its own kinematic and dynamic constraints. In our application, the model of the local environment is provided to the vehicle by an on-board perception system (cameras and internal sensors). At any time instant, the vehicle obtains information about the local structure of the roadway (number of lanes, geometrical characteristics of the lanes and road marks), the moving obstacles around, the current position of the vehicle and its internal state (velocity, acceleration and steering angle). Different kinds of obstacles are considered: other vehicles, motorcycles, pedestrians, animals, etc. The framework of this research work is the French demonstrator Pro-Lab II, part of the European programme on transport Prometheus. The purpose of Pro-Lab II is to design, integrate and demonstrate an 'electronic copilot' to assist the human driver in various traffic situations.