Real-time path planning for a robot to track a fast moving target based on improved Glasius bio-inspired neural networks

Abstract

Path planning from the initial location to reach the target location has received considerable attentions. There are still some challenges in real-time path planning for tracking a fast moving target. In the application of tracking a fast moving target, the robot must move rapidly and correctly to follow the target. It requires an algorithm that enables real-time path planning in the fast changing environment. The Glasius Bio-inspired Neural Network (GBNN) model has been reported inefficient for the real-time path planning in the fast changing environments because its dynamic performance lags behind the fast environmental changes. This study puts forward several improvements for the GBNN model to improve its dynamic performance. An alternate weight function for GBNN model is also studied. It shows that all the proposed GBNN models can be parameter-fixed when applied in a specific case. Through theoretical analysis and comparative simulations, the improved models are stable and feasible in the real-time path planning in a rapidly changing environment.