基于FPGA的轨道式移动机器人偏摆力矩跟随控制系统设计

Design of FPGA based yaw moment following control system for orbital mobile robot

摘要:

在高速轨道运行环境中,受轨道谱激励、负载突变等多源扰动耦合作用,轨道式移动机器人易呈现显著的偏摆力矩非线性波动,导致控制存在偏差。为此,设计了基于FPGA的轨道式移动机器人偏摆力矩跟随控制系统。系统以XilinxKintex-7FPGA为核心,构建了集多维度运动参数感知、硬件并行运算与高精度数模转换于一体的控制架构。通过将三次样条插值与参数搜索过程硬件化,系统能够快速遍历参数空间并嵌入机械约束,从而兼顾轨迹平滑性与实时性要求。通过构建总作用力矩与偏摆力矩的耦合修正模型,结合实时解算与PID控制,系统可动态输出补偿指令,从而有效解决了偏摆力矩的非线性波动问题,实现微秒级力矩补偿与实时跟随控制。实验结果表明,在偏心负载突变场景下,该系统能将偏摆力矩波动幅度抑制在0.19–0.32N·m范围内,侧导轮调节臂角度跟踪曲线与理论指标一致,即使在0.6N·m峰值力矩下仍能稳定跟随,且偏摆力矩峰值为50N·m时,偏摆力矩抗干扰恢复时间仅为13s,验证了FPGA控制在强干扰场景下的鲁棒性。

Abstract:

In the high-speed rail operating environment, due to the coupling effect of multiple sources of disturbances such as track spectrum excitation and load mutation, the rail type mobile robot is prone to significant nonlinear fluctuations in yaw moment, leading to control deviations. For this purpose, a trajectory based mobile robot yaw moment following control system based on FPGA was designed. The system is based on Xilinx Kintex-7 FPGA and has built a control architecture that integrates multi-dimensional motion parameter sensing, hardware parallel computing, and high-precision analog-to-digital conversion. By hardware implementing cubic spline interpolation and parameter search process, the system can quickly traverse the parameter space and embed mechanical constraints, thus balancing trajectory smoothness and real-time requirements. By constructing a coupled correction model for the total applied torque and deflection torque, combined with real-time calculation and PID control, the system can dynamically output compensation instructions, effectively solving the nonlinear fluctuation problem of deflection torque and achieving microsecond level torque compensation and real-time follow-up control. The experimental results show that in the scenario of sudden changes in eccentric load, the system can suppress the fluctuation amplitude of the deflection torque within the range of 0.19-0.32 N·m. The angle tracking curve of the side guide wheel adjustment arm is consistent with the theoretical index, and it can still stably follow even under the peak torque of 0.6 N·m. When the peak deflection torque is 50 N·m, the anti-interference recovery time of the deflection torque is only 13 seconds, which verifies the robustness of FPGA control in strong disturbance scenarios.