Towards a Hybrid Real/Virtual Simulation of Autonomous Vehicles for Critical Scenarios
Franck Gechter, Baudouin Dafflon, Pablo Gruer, Abderrafìâa Koukam · 2014
Abstract—Developing control and perception algorithms for au-tonomous vehicle is a time-consuming activity if one performs it directly on vehicles hardware level. Moreover, some test cases are hard to reproduce. For this reason, many laboratories and companies are generally using simulation tools. The goal of these tools is to benefit from a testing environment as close as possible to reality and able to reproduce specific testing cases. The main problem with standard simulation tools is that they may not accurately represent real conditions. In order to increase the quality of the simulations, hardware is generally introduced in the loop. The goal of this paper is to present a ”work-in-progress” adaptation of IRTES/SeT-Lab simulation tool named VIVUS so as to be able to introduce hybrid simulation which consists in both introducing hardware in the simulation loop and/or software simulation in the hardware experimental loop. Keywords–hybrid simulation; autonomous vehicle algorithms; sensors simulation; augmented reality I.