Title: Optimising inverse kinematics algorithm for an indigenous vision-based feeding serial robot using particle swarm optimisation and hybrid genetic algorithm: a comparison

Authors: Priyam A. Parikh; Reena Trivedi; Keyur D. Joshi

Addresses: Institute of Design, Nirma University, Ahmedabad, India ' Institute of Technology, Nirma University, Ahmedabad, India ' School of Engineering and Applied Sciences, Ahmedabad University, India

Abstract: This paper aims to provide an optimal inverse kinematics solution for an indigenous 6 DoF feeding robot using evolutionary algorithms such as C-PSO and H-GA. Here, a case of a vision-based 3D printed serial manipulator is taken, which helps patients with meal consumption. A robotic arm passes through many intermediate points in its entire trajectory, which might create a positional error in Euclidean-space. The higher positional error can lead the robot's end-effector to the incorrect destination. To overcome this problem, we have provided a methodology that would help to perform IK at every intermediate point using C-PSO and H-GA. To efficiently solve the problem of positional error, the IK was optimised using C-PSO and H-GA, which gave a mean PE of 4.95% and 3.78% respectively. Finally, the PE, obtained from C-PSO and H-GA were compared and plotted in 2D line and 3D surface plots respectively.

Keywords: feeding robot; evolutionary algorithms; Euclidean distance error; inverse kinematics; particle swarm optimisation; genetic algorithm.

DOI: 10.1504/IJAMECHS.2023.131332

International Journal of Advanced Mechatronic Systems, 2023 Vol.10 No.2, pp.88 - 101

Received: 17 Oct 2022
Accepted: 10 Jan 2023
Published online: 06 Jun 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article