Implementation of a modified hamiltonian algorithm for control allocation
Paper in proceeding, 2017

This paper is motivated by the problem of terminal understeer of a road vehicle. A relevant scenario is where a vehicle enters a curve at excessive speed and the control objective is to apply automatic chassis control to prevent the vehicle from drifting out of the lane. In a previous study, the optimization problem was formulated as the minimization of maximum path off-tracking; the optimal response approximates to that of a friction-limited particle model, leading to a parabolic path reference (PPR) with the target mass-centre acceleration vector fixed in the global frame. The present paper considers in greater detail how the acceleration target is realized on an experimental vehicle. The approach is based on a recently developed model-based control method – the Modified Hamiltonian Algorithm (MHA). Simulations with a full vehicle (CarMaker) model are carried out, and test-track experiments are conducted to test the controller and validate the simulations. Results show that transient effects such as actuator delays and rate limits can have a significant effect on performance. On the other hand, the controller appears robust to factors such as tyre wear and uncertainty in vehicle parameters, and closely approximates the particle reference once these transient effects are accounted for. Since the control architecture is adaptive to track a changing acceleration reference, these conclusions are likely to be valid across a wider range of scenarios.

Author

Y. Gao

University of Lincoln

Tim Gordon

University of Lincoln

Mathias R Lidberg

Chalmers, Applied Mechanics, Vehicle Safety

Vehicle and Traffic Safety Centre at Chalmers

Proceedings of the 13th International Symposium on dvanced Vehicle Control, AVEC 2016;

Vol. 2017 157-162
978-131526528-5 (ISBN)

13th International Symposium on Advanced Vehicle Control, AVEC 2016
München, Germany,

Areas of Advance

Transport

Subject Categories

Robotics

Control Engineering

More information

Latest update

2/19/2021