Robust trajectory tracking control of lunar rovers considering wheel-terrain slip

This study addresses wheel-terrain slip in lunar rover navigation by proposing a robust control algorithm with slip compensation. An enhanced kinematic model incorporating slip velocity components is established, and error dynamics are derived in the body-fixed frame. Building upon these equations, a stepwise procedure is proposed to design a robust controller, with demonstrated asymptotic convergence properties for tracking errors. The algorithm is evaluated through simulations involving nonlinear-curvature paths on inclined terrain. Results show significant tracking improvement: compared to an uncompensated controller, the root mean square error (RMSE) is reduced by 90.6 % and 85.4 %; relative to an existing slip-compensated robust method, reductions of 61.3 % and 73.6 % are achieved. Furthermore, chattering in angular velocity is reduced by 73.9 % and 77.9 %, indicating improved stability. Additional tests on rubble-covered zone confirm robustness against dynamic disturbances and inaccurate boundary estimation. The algorithm maintains stability even under underestimated disturbances, though a trade-off between accuracy and chattering is observed when increasing boundary estimates. Real-time physical deployment over 10,000 control cycles demonstrates a worst-case execution time of 0.862 ms, occupying only 0.9 % of a typical 100 ms control loop period, confirming sufficient computational margin.