论文部分内容阅读
通过自适应动态规划研究自由漂浮空间双臂机器人运动的最优控制问题.针对空间双臂机器人的非完整性运动,采用自适应动态规划(Adaptive Dynamic Programming,ADP)方法求解其最优控制问题.根据多体动力学理论,推导了载体位置、姿态均无控制条件下,双臂空间机器人满足的系统动量守恒关系的非完整约束方程,并将其转化为控制系统的状态方程,从而将双臂空间机器人的非完整运动规划问题转化为对非线性系统的控制问题.文中根据自适应动态规划网络结构,利用神经网络来近似性能指标函数,进而用龙格库塔法求解状态变量.并给出了适合该类问题的一种效用函数具体表达式,保证了空间双臂机器人到达期望位置后不再继续运动.实现了对空间双臂机器人非完整运动规划的最优控制.数值仿真实验验证了ADP对求解空间双臂机器人非完整运动规划最优控制问题的有效性.
The optimal control of two-arm robots in free-floating space is studied through adaptive dynamic programming.In order to solve the non-integral motion of two-arm space robots, an adaptive dynamic programming (ADP) method is used to solve the optimal control problem. According to the multi-body dynamics theory, the nonholonomic constrained equations of the system’s momentum conservation relationship satisfied by the two-arm space robot are derived under the condition of no control of the position and attitude of the carrier, and then transformed into the state equation of the control system, The problem of nonholonomic motion planning in space robot is transformed into the control problem of nonlinear system. According to adaptive dynamic programming network structure, the neural network is used to approximate the performance index function, and then the state variable is solved by Runge-Kutta method. A concrete expression of utility function suitable for this kind of problem ensures that the two-arm robot in the space will not continue its motion when it reaches the desired position, which realizes the optimal control of the nonholonomic motion planning of space robot with two arms. Effectiveness of ADP in Solving Optimal Control of Nonholonomic Motion Planning of Two - Arm Robots.