Optimized skill knowledge transfer model using hybrid Chicken Swarm plus Deer Hunting Optimization for human to robot interaction

Abstract

Robots are more competent for progressing knowledge and learning new tasks that are of demanding interest. Service robots need trouble-free programming techniques facilitating the inexperienced human user to easily incorporate motion and perception skills or complex problem-solving strategies. The modern advancement in knowledge offers the most potent solution to this problem, enabling robots to incite enhanced skill knowledge transferability. This paper tempts to develop an efficient skill knowledge transfer between humans and computers using Transfer Expert Reinforcement Learning. Here, the movement of the robotic arm is considered which is solved with the assistance of one of the basic machine learning paradigms termed as Reinforcement Learning. As an improvement, the action features of Reinforcement Learning are optimized by a hybrid meta-heuristic algorithm with the integration of the Deer Hunting Optimization Algorithm and Chicken Swarm Optimization termed as Chicken Swarm-plus Deer Hunting Optimization Algorithm. Moreover, the Artificial Neural Network plays a major role here to determine the desired movement based on the input kinematic movements. The main objective of optimized Reinforcement Learning is to maximize the reward, which in turn minimizes the error difference between the desired and the predicted movement. The experimental analysis of the Chicken Swarm-plus Deer Hunting Optimization Algorithm based Reinforcement Learning over the conventional and other heuristic-based Reinforcement Learning proves the effective performance of the proposed model.