【作者】 孙玉德；

【导师】 梅江平；

【作者基本信息】 天津大学 ， 机械工程， 2017， 硕士

【摘要】 本文密切结合MD-1200型码垛机器人在PTP模式下的应用,系统研究了其以低能耗为目标的轨迹优化方法,包括运动学和动力学建模、轨迹规划及其优化,并在该机器人上开展验证实验。论文取得的成果如下:在运动学方面,基于对MD-1200型码垛机器人的结构分析,依据D-H法建立了该机器人连杆坐标系,推导出其正逆运动学模型。给出该机器人尺度参数,运用MATLAB得到其末端可达工作空间以及横截面形状。在动力学方面,利用Kane方程求解出各关节驱动力矩,建立码垛机器人不含摩擦项的逆动力学模型。基于Stribeck摩擦模型,得到各关节摩擦力矩。从而建立了MD-1200型码垛机器人完备动力学模型。在轨迹规划方面,以运动学模型为基础,分别在操作空间和关节空间,基于3-4-5次多项式运动规律和5次非均匀B样条运动规律建立操作空间运动参数(位移、速度以及加速度)与关节空间运动参数(角位移、角速度以及角加速度)的函数映射。通过逆动力学模型,求得各关节力矩。评价码垛机器人在两种轨迹规划方法下的动态性能、摩擦能耗以及单位时间内关节平均能耗,并且考虑了末端最大加速度和运动周期对于关节能耗的影响。仿真结果表明5次非均匀B样条运动规律在轨迹规划方面更具优越性。在轨迹优化方面,基于加权系数法,提出一种综合考虑码垛机器人能耗和动态性能的评价指标。结合5次非均匀B样条运动规律的轨迹规划方法,运用遗传算法针对评价指标实现轨迹优化,得到轨迹分段时间的最优解。比较优化前后码垛机器人在“门”字型轨迹下的动态性能、摩擦能耗以及关节单位时间内的平均能耗情况,结果表明优化后的轨迹对于提升码垛机器人的动态性能和能耗水平具有显著作用。在实验方面,搭建了MD-1200型码垛机器人实验平台,对以上所研究的轨迹规划以及优化方法进行了实验验证。证明了本文所提出的以低能耗为目标的轨迹优化方法在保证码垛机器人动态性能以及改善能耗方面的有效性。本文的研究内容对于丰富码垛机器人轨迹优化方法的研究,实现其在工程应用中的低能耗作业具有重要意义。更多还原

【Abstract】 This paper,closely based on the application of MD-1200 palletizing robot in point-to-point mode,had presented the method to optimize the trajectory of palletizing robot based on energy optimization,including kinematic and dynamic modelings,trajectory planning and optimization.Then,the experiment on MD-1200 palletizing robot was achieved.The main contributions were concluded as following:In terms of kinematics,the structure of MD-1200 palletizing robot was analyzed.According to D-H method,the robot connecting rod coordinate system was established and its direct and inverse kinematics model was deduced.Based on the dimension parameters,the workspace and cross section shape were obtained in MATLABIn the aspect of dynamics,Kane function was used to derive the joint driving torque.And palletizing robot inverse dynamic model without friction torque was established.Based on stribeck friction model,the friction torque of each joint was got.On this basis,the MD-1200 palletizing robot complete dynamic model was achieved.In the trajectory planning,based on the kinematic model,the 3-4-5 polynomial motion law had achieved the mapping relation between operating space kinematic parameters(displacement,velocity,acceleration)and joint space kinematic parameters(angular displacement,angular velocity,angular acceleration)in operating space.And using quintic non-uniform B-splint motion law,the mapping relation between joint space kinematic parameters and operating space kinematic parameters had been achieved in joint space.According to the inverse dynamic model,torque of each joint was obtained.The dynamic performance,friction energy and average energy of joins in unit time of the palletizing robot under the two trajectory planning methods were evaluated.And the maximum acceleration of end-effecter and period of motion and their effects on the energy of joints was considered.The result of simulation shows that quintic non-uniform B-splinemotion law has more advantages in terms of trajectory planning.Based on the weighting factor method,a multi-objective evaluation index including energy and dynamic performance was proposed.The optimization solution was obtained by using the genetic algorithm combined with quintic non-uniform B-spline motion law to optimize the trajectory.By comparing the dynamic performance,friction energy and average energy of joins in unit time of the palletizing robot before and after optimization under the "door" trajectory,the result shows that the optimized trajectory has significant effect on improving the dynamic performance and energy consumption.In the experiment,the MD-1200 palletizing robot experiment platform was established to verify the researches on trajectory planning and trajectory optimization.The method,which was aimed at energy trajectory optimization,was proved to be effective in ensuring the palletizing robot dynamic performance and improving the energy of joints.The research of this paper is of important significance in enriching the research of palletizing robot trajectory optimization and achieving lower energy operation in engineering application.更多还原