Inverse kinematics calculation method of banana picking robot based on Newton iteration method

Abstract

The invention discloses a banana picking robot inverse kinematics calculation method based on a Newton iteration method. The banana picking robot inverse kinematics calculation method comprises the following steps that according to a mechanical structure of a picking robot, a forward kinematics model of the position and posture of a banana fruit stem clamping piece in a picking actuator and all degrees of freedom is established; and an inverse kinematics solution meeting the set precision is obtained based on a Newton iteration algorithm according to the target position and posture of the banana fruit stem clamping piece. According to the method, inverse kinematics analytic solutions of a base rotation angle theta and a tail end rotation angle are obtained according to a forward kinematics model and a geometrical relationship of the picking robot, then an equation set met by horizontal displacement x and vertical displacement z is constructed through the forward kinematics model of the picking robot, and an inverse matrix of a Jacobian matrix of the equation set is obtained; and then inverse kinematics solutions of the horizontal displacement x and the vertical displacement z meeting the set precision are obtained according to the target position and posture of the fruit stem clamping mechanism through the Newton iteration algorithm, the precision is high, and the real-time performance is good.

Description

technical field [0001] The invention relates to the technical field of robot picking, in particular to an inverse kinematics calculation method for a banana picking robot based on the Newton iteration method. Background technique [0002] Banana is a characteristic fruit and economic crop in Lingnan. The mass of banana bunches in the right harvest period is about 30kg, and the height of the cutting part of the fruit stem exceeds 2m. At present, the picking of banana bunches is mainly done manually, which is completed by the cooperation of multiple people. The labor intensity is high, the labor cost is high, and the picking efficiency is low, which seriously restricts the industrial development and economic benefits of bananas. Robots can realize the mechanization and Intelligentization can not only improve picking efficiency and reduce costs, but also solve the problem of agricultural labor shortage. The picking robot first plans the movement trajectory of the banana handle...

AI Technical Summary

Problems solved by technology

However, the neural network structure is not easy to determine, there are many training parameters, the training sample size is large and difficult to obtain, and the network training time is long, which requires high hardware configuration of the robot control system; reinforcement learning also requires a large number of data samples, and the training process is relatively slow. Long; the iterative optimization process of the particle swarm optimization algorithm is complex, and the total number of particles is difficult to determine; the genetic algorithm is prone to premature in solving the inverse kinematics process, and the local search ability is poor, and it may not be possible to obtain an inverse kinematics solution that meets a given accuracy every time

Images