面向复杂地形场景的四轮独立转向重载机器人端到端控制OA

To address the limited generalization of traditional control methods for four-wheel independent steering and driving(4WISD)mobile robots in complex terrains,caused by their reliance on idealized constraints and para-meter tuning as well as their susceptibility to non-ideal wheel-ground interactions,an end-to-end deep reinforce-ment learning low-level control policy without explicit model-based computation is developed.Based on proximal policy optimization,the method integrates curriculum learning and physics-based domain randomization to con-struct a multi-terrain and multi-disturbance training process.With a structured state-action formulation and phys-ically motivated reward shaping,the policy directly maps proprioceptive observations to four-wheel steering-angle and wheel-speed commands,enabling coordinated multi-wheel control and slip suppression.Experiments in the high-fidelity Isaac Sim simulator compare the proposed policy with PID-LS and MPC-LS cascaded baselines under rough undulating terrains and low-friction flat terrains.Results show that the proposed policy exhibits smaller perform-ance fluctuations across conditions,achieves more accurate linear-velocity tracking on rough terrains,and main-tains low online computational cost,with a single-step latency of 0.304 ms.Real-world validation on an outdoor flat concrete surface further demonstrates stable online execution through a ROS-PLC control pipeline,with velocity root mean square error(RMSE)values of 0.169 m/s,0.213 m/s,and 0.116 rad/s for the three velocity components.The policy inference latency is 0.62 ms,while sensor synchronization and state construction take 56.15 ms on aver-age.These results verify the feasibility of end-to-end data-driven low-level control for 4WISD platforms without pre-cise modeling,offering an innovative approach to motion control of high-DOF wheeled mobile robots in complex en-vironments.