An interactive demo of autonomous vehicle navigation was presented at the “Sergiu T. Chiriacescu” Aula, during an even, called “IESC si companiile” (Electrical Engineering Faculty and the Companies), organized by the Transilvania University of Brasov. The Audi Model Car vehicle highlighted the entire autonomous vehicle processing pipeline, from perception to higher level algorithms.
The main challenges were indoor localization and navigation, dynamic obstacle avoidance, as well as difficult surface and lighting conditions. The vehicle had to navigate around a global trajectory and had to be capable of avoiding obstacles with minimum deviation from the global trajectory, without the aid of visual clues for localization and control purposes.
The sensing part of the robot is produced by ultrasonic sensors, which are projected onto a grid map that acts as input for the local planner module. The local planner computes the local trajectory, that has the role of providing a navigable path around detected obstacles, with deviation from the global trajectory being as small as possible.
The controller module of the vehicle is implemented as a Model Predictive Controller path tracking algorithm, which aims at minimizing the cross-track error and angle error between the vehicle and the local trajectory provided by the local planner module.
Check a demo in the video below.