A Bio-Inspired Neural Network Approach to Robot Navigation and Mapping with Nature-Inspired Algorithms

Nature-inspired algorithms have been successfully applied to autonomous robot path planning, vision and mapping. However, concurrent path planning and mapping with replanning feature is still a challenge for autonomous robot navigation. In this paper, a new framework in light of the replanning-based methodology of concurrent mapping and path planning is proposed. It initially performs global path planning through a developed Gravitational Search Algorithm (GSA) to generate a global trajectory. The surrounding environment can then be described through a monocular framework and transformed into occupancy grid maps (OGM) for autonomous robot path planning. With updated moving obstacles and road conditions, the robot can replan the trajectory with the GSA based on the updated map. Local trajectory in the vicinity of the obstacles is generated by a developed bio-inspired neural network (BNN) method integrated with speed profile mechanism, and safe border patrolling waypoints. Simulation and comparative studies demonstrate the effectiveness and robustness of the proposed model.