摘要
为了解决SCARA机械臂因目标位姿变化而引起抓取任务失败问题,给出了一种基于机器视觉目标检测与定位的机械臂控制方法。通过双目相机成像模型实现坐标系之间的转换,利用计算机视觉识别及三维匹配方法,达到对目标物体定位的目的,从而确定机械臂末端执行器的期望位置。结合使用D-H参数法建立机械臂的运动学模型,通过运动学方程的逆解,得到末端执行器到达期望位置时机械臂各驱动关节的变量参数,然后对其进行轨迹规划仿真,得到了连续且平滑的各关节位移、速度、加速度变化规律曲线,同时使用SCARA机械臂进行装配实验。实验结果表明该系统能够准确、稳定地把物体放到目标位置,完成装配任务,从而为开发机械臂运动控制系统提供重要的参考。
A control method for SCARA manipulator is presented on the basis of the detection and location of machine vision to solve the robotic gripping failure caused by the change of target pose.The coordinate system is transformed by employing binocular camera imaging model,and the location of target object is realized through computer vision recognition and three-dimensional matching method to determine the expected position for the end-effector of the manipulator.The variable parameters of each driving joint while the end-effector reaching the expected position is obtained by establishing kinematics model though D-H parametric method and solving the inverse kinematics equation.Then,a set of continuous and smooth curves of joint displacement,velocity and acceleration are obtained by simulation.Meanwhile,the assembly test on the SCARA manipulator is carried out.The results indicate that the system is able to move the object to the target position and complete the assembly mission accurately and stably,which provides important reference for the development of the motion control system for manipulators.
作者
李旭
裴旭明
刘迪
李浩
LI Xu;PEI Xuming;LIU Di;LI Hao(Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment,Zhengzhou University of Light Industry,Zhengzhou 450002,China)
出处
《机械》
2020年第2期7-13,58,共8页
Machinery
基金
国家自然科学基金资助项目(51775517)
河南省科技攻关项目(102102210134)。