Sensor-Based Control with Digital Maps Association for Global Navigation: A Real Application for Autonomous Vehicles

This paper presents a sensor-based control strategy applied in the global navigation of autonomous vehicles in urban environments. Typically, sensor-based control performs local navigation tasks regarding some features perceived from the environment. However, when there is more than one possibility to go, like in road intersection, the vehicle control fails to accomplish its global navigation. In order to solve this problem, we propose the vehicle global navigation based on a topological representation of the environment using only digital maps and low cost sensors. The approach was developed for two main tasks: road following and road intersection maneuvers. The final solution was completely implemented in a real autonomous car and tested in a challenge circuit, showing the viability of our solution.