基于大数据聚类的机器人步态控制系统设计
2022,30(3):95-99
摘要：为提高机器人关节的运动灵活性，使其具有更快、更稳的步行能力，设计基于大数据聚类的机器人步态控制系统。结合外接主电源与传感器设备，为主处理器元件、舵机控制板提供足量的传输电信号，完成机器人步态控制系统硬件设计。定义关键的关节节点，通过建立适应度函数的方式，实现基于大数据聚类的步态节点安排。分别从支撑腿运动规划、摆动腿运动规划两方面着手，收集大量的步态运动信息，再联合已知函数条件，对机器人关节角度进行求解，完成机器人行进步态的规划与处理，实现基于大数据聚类机器人步态控制系统的应用。实验将所得关节弯曲次数、角度值与理想数值对比可知，本文系统位姿标定情况下关节弯曲次数值大、角度值小，机器人关节的运动灵活性水平高，具有相对稳定的步行运动能力。
关键词：大数据聚类；机器人；步态控制；适应度函数；支撑腿；摆动腿；关节角度
Design of Robot Gait Control System Based on Big Data Clustering
Abstract：In order to improve the movement flexibility of the robot joints and make it have a faster and more stable walking ability, a robot gait control system based on big data clustering is designed. Combined with the external main power supply and sensor equipment, the main processor element and the steering gear control board provide sufficient transmission electrical signals to complete the hardware design of the robot gait control system. Define the key joint nodes, and realize the gait node arrangement based on big data clustering by establishing the fitness function. Starting from the two aspects of supporting leg motion planning and swinging leg motion planning, a large amount of gait motion information is collected, and then the known function conditions are combined to solve the robot joint angles to complete the planning and processing of the robot’s progress state, and realize the realization of large-scale Application of data clustering robot gait control system. The experiment compares the obtained joint bending times and angle values with the ideal values. It can be seen that the joint bending times and angle values are large in the case of the posture calibration of the system in this paper. The robot joints have a high level of mobility and relatively stable walking ability.
Key words：big data clustering; robot; gait control; fitness function; support leg; swing leg; joint angle;
收稿日期：2021-08-23
基金项目：辽宁省中央引导地方科技发展资金沈抚示范区本级科技计划项目: 基于过程神经网络的移动机器人RTK-GPS导航定位研究，编号：2021JH6/10500137
