Zonotopic-tube-based LPV motion planner for safety coordination of autonomous vehicles

© 2023 The Authors. This is an open access article under the CC BY-NC-ND license. Peer review under responsibility of International Federation of Automatic Control.
In this paper an optimization-based solution to the collision avoidance challenge for autonomous vehicles is proposed. The presented approach consists in an online motion planner designed to define a feasible and efficient path which implicitly guarantees safety manoeuvres in dynamic surroundings. The fact of considering moving obstacles inside the motion planner increases the complexity of the problem while forces it to be executed more frequently as others. To reduce its computational complexity, this approach proposes a two stages translation of the commonly used non-linear optimization-based structure into a QP formulation which can be easily solved. The first stage is based on the use of LPV matrices in the dynamic constraints of the vehicle. The second stage consists in computing linear expressions by set propagation to obtain the set of permitted inputs and reachable states which guarantee safety conditions.
This work has been co-financed by the Spanish State Research Agency (AEI) through the projects SaCoAV (ref. MINECO PID2020-114244RB-I00) and MASHED (TED2021-129927B-I00).
Peer Reviewed
Postprint (published version)