Robust trajectory planning for robotic communications under fading channels

Published in Ubiquitous Networking, Third International Symposium, UNet 2017, Casablanca, Morocco, May 9-12, 2017, Revised Selected Papers, E. Sabir, A. García Armada, M. Ghogho, M. Debbah (Eds.), Computer Communication Networks and Telecommunications Series, vol. 10542, p. 450-460, Springer International Publishing, 2017 International audience We consider a new problem of robust trajectory planning for robots that have a physical destination and a communication constraint. Specifically, the robot or automatic vehicle must move from a given starting point to a target point while uploading/downloading a given amount of data within a given time, while accounting for the energy cost and the time taken to download. However, this trajectory is assumed to be planned in advance (e.g., because online computation cannot be performed). Due to wireless channel fluctuations, it is essential for the planned trajectory to be robust to packet losses and meet the communication target with a sufficiently high probability. This optimization problem contrasts with the classical mobile communications paradigm in which communication aspects are assumed to be independent from the motion aspects. This setup is formalized here and leads us to determining non-trivial trajectories for the mobile, which are highlighted in the numerical result.