基于神经网络的机器人操作手IKP精确求解

详细信息

AN ACCURATE SOLUTION TO THE INVERSE KINEMATIC PROBLEM OF A ROBOT MANIPULATOR BASED ON THE NEURAL NETWORK

Mechanism Electronic Engineering Specialty, Harbin Institute of Technology 150001

摘要: 结合位置正解模型，利用BP网络求解了机器人逆运动学问题（IKP）．为提高求解结果精度，采用迭代计算进行误差补偿，计算结果表明，该法迭代次数少，计算精度高且计算速度接近机器人实时控制的要求．
- 机器人操作手 /
- 逆运动学问题（IKP） /
- 神经网络 /
- 误差补偿
Abstract: In this paper, IKP of the robot manipulator is solved by using BP neural network and the forward kinematic model. To improve the accuracy of the solution, an iterative approach is used to compensate for the offset error. Numerical results have shown that the accurate solutions can be obtained by performing only a few iteration steps and the computation speed can meet the requirements for the robot's real time control system.
- robot manipulator /
- inverse kinematics /
- neural network /
- error compensation

[1]	Y F Lou, P Brunn. A hybrid artificial neural network inverse kinematic solution for accurate robot path control. Proc Instn Mech Engrs 1999,213(1):23-32.
[2]	L Acosta, G N Marichal, et al. A robotic system based on neural network controllers. Artificial Intrlligence in Engineering. 1999(13):393-398.
[3]	Simon X Yang, Max Meng. An efficient neural network method for real-time motion planning with safety consideration. Robotics and Autonomous Systems,2000(32):115-128.
[4]	Bekir Karlik, Serkan Aydin. An improved approach to the solution of inverse kinematics problems for robot manipulators. Engineering Applications of Artificial Intelligence,2000(13):159-164.
[5]	D T Pham, Sahin Yildirim. Control of the trajectory of a planar robot using recurrent hybrid networks. International Journal of Machine Tools & Manufacture,1999(39):415-429.
[6]	张际先,宓霞.神经网络及其在工程中的应用.北京:机械工业出版社,1996.
[7]	付京逊,杨静宇.机器人学:控制、传感技术、视觉、智能.北京:中国科学技术出版社,1989