Multi-criteria trajectory optimization for autonomous vehicles

In the last few years much effort has been made towards more autonomous vehicles and fuel consumption reduction. This article deals with the issue trajectory optimization of unmanned terrestrial vehicles so as to reduce consumption, travel time or to improve comfort. Main focuses are set on testing different criteria and the possibility of using a genetic algorithm to improve the potential field methods (Ge and Cui (2002) and Melchior et al. (2003)). The main idea of this article is that potential field methods could be improved by smartly placing intermediate attractive points in the field. It brings two improvements to the potential field method: the generation of an optimal path in the environment, and the generation of a correlated optimal motion. In the first two parts of this article the issue at stake is briefly described along with the different criteria and methods used, then simulations will be presented of the potential field method and its combination with a genetic algorithm. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.