Abstract: Aiming at the common failure in the drive motor of the Four-Wheel Independent Drive and Steering (4WID-4WIS) type of port heavy-duty Automated Guided Vehicle(AGV),a fault-tolerant control strategy with a three-layer control structure was proposed to ensure the steering stability under the circumstances of drive motor failure.In Upper control module,a Model Predictive Controller(MPC) and a PI vehicle speed following controller were consisted to realize the trajectory tracking performance of port AGV.Then,the middle controller was designed as a sliding mode controller with Radial Basis Function (RBF) neural network robust to adjust the yaw rate and the center of mass sideslip angle,which could be used to calculate the optimal additional yaw moment.The lower control module provided a failure distribution strategy to redistribute the torque.In addition,an CarMaker test platform was built,and the fault-tolerant control method verified its the validity and superiority by the simulation experiments.

Key words: port heavy duty AGV, model predictive controller, PI speed follow controller, radial basis function network neural network, sliding mode control, failure allocation

摘要： 针对四轮独立驱动与转向(4WID-4WIS)型港口重载自动导引车(AGV)常见的驱动电机失效情况,提出了三层控制结构的容错控制策略来进行转向稳定性容错控制。上层控制模块设计为模型预测控制器(MPC)与PI车速跟随控制器,实现港口AGV的路径跟踪;中层控制模块设计为横摆角速度、质心侧偏角RBF神经网络鲁棒滑模控制器,用来计算出最佳附加横摆力矩;下层控制模块设计为失效分配策略,对力矩进行重新分配。最后,搭建了CarMaker测试平台,通过实验验证了容错控制策略的有效性与优越性。

关键词: 港口重载AGV, 模型预测控制器, PI车速跟随控制器, RBF神经网络, 滑模控制, 失效分配