摘要
为提高凿岩机器人钻臂末端(钎头)的定位精度,在利用粒子群优化(PSO)算法对关节变量误差进行补偿时,存在收敛速度慢、容易过早陷入局部最优解等问题,为此,提出一种交叉精英反向粒子群优化算法(CEOPSO)并给出算法流程。针对影响误差的两个主要因素,采用五参数D-H方法建立钻臂的参数误差模型,在形变关节后引入一个虚拟关节,推导出钻臂的形变误差模型。将交叉算子引入到EOPSO算法中,同时进行自适应惯性权重和交叉概率参数控制,不仅维持了粒子个体与最优解之间的信息交换,而且增加了粒子个体之间的信息交换。对比仿真结果表明,在误差补偿控制过程中,CEOPSO算法具有更优越的最优关节补偿值搜索收敛速度和求解稳定性,提高了凿岩机器人钻臂的定位控制性能。
In order to improve the positioning accuracy of rock drilling robotic drilling arm,the particle swarm optimization（ PSO） algorithm is often used to compensate the joint variable error of rock drilling robotic drilling arm,but there are some problems,such as low convergence speed,tending to be trapped in local optimal solution,etc. In order to solve these problems,a crossover elite opposition-based particle swarm optimization（ CEOPSO）algorithm is presented and the algorithm flow is given in this paper. Aiming at the two main error factors which are parameter error and deformation error,the parameter error model of drilling arm is established by modified D-H method,and the deformation error model of the drilling arm is derived by introducing a virtual joint into the deformed joint. The crossover operator is introduced into EOPSO. The adaptive inertia weight and the crossover probability parameter control technology are adopted. On the basis of maintaining the information exchange between the individual and the optimal solution,the global searching ability of the algorithm and the positioning efficiency of drilling arm are improved by increasing the information exchange between the individual particles. Simulation results show that the CEOPSO has better optimal joint compensation value searching convergence speed and solving stability than those of PSO and EOPSO. The positioning and control performance of rock drilling robotic drilling arm can be improved effectively.
作者
黄开启
魏文彬
陈荣华
丁问司
Huang Kaiqi;Wei Wenbin;Chen Ronghua;Ding Wensi(School of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Jiangxi Ganzhou 341000, China;School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, China)
出处
《机械科学与技术》
CSCD
北大核心
2018年第7期1005-1012,共8页
Mechanical Science and Technology for Aerospace Engineering
基金
国家自然科学基金项目(11272122)
广东省部产学研重大项目(2012A090300011)
江西省科技厅对外合作重点项目(20123BBE50085)资助
关键词
凿岩机器人
钻臂
误差补偿
粒子群优化
精英反向学习
交叉算子
rock-drilling robot
drilling arm
error compensation
particle swarm optimization （PSO）
elite opposition based learning
crossover operator
