论文部分内容阅读
针对多关节果实收获机器人难以获得精确控制模型以及控制系统的抖振问题,提出基于遗传算法实时动态调整滑模参数的控制策略,设计并制作了基于STM32微控制器和AS5045位置反馈模块及CAN总线通信模块的关节控制系统仿真和试验平台,分别在空载与负载情况下进行了关节电机位置响应试验。结果表明,采用遗传算法动态调整滑模控制器参数能够提高关节控制系统位置跟踪的响应速度,减少外界干扰和负载变化引起的控制系统抖振的幅度与持续时间,具有较强的鲁棒性。由空载和负载试验结果可知,关节6实际试验控制系统的位置跟踪响应时间比理论仿真试验增加了0.5 s,负载时控制系统的位置跟踪响应时间比空载时增加了0.3 s,但负载对系统精度和超调量并无明显影响,系统具有良好的控制效果。
Aiming at the difficult control model of multi-joint fruit-harvesting robots and the chattering of control system, a control strategy of real-time dynamic adjustment of sliding-mode parameters based on genetic algorithm is proposed. Based on STM32 microcontroller and AS5045 position feedback module and CAN bus Communication module joint control system simulation and test platform, respectively, under no-load and load joint motor position response test. The results show that using genetic algorithm to dynamically adjust the parameters of sliding mode controller can improve the response speed of position tracking of the joint control system and reduce the amplitude and duration of the chattering caused by external disturbance and load changes and has strong robustness. According to the no-load and load test results, the position tracking response time of the actual test control system of joint 6 is 0.5 s longer than that of the theoretical simulation test, and the position tracking response time of the control system is increased by 0.3 s when the load is compared with the no-load test. However, System accuracy and overshoot did not significantly affect the system has a good control effect.