Abstract

A hybrid path planning algorithm based on membrane pseudo-bacterial potential field (MemPBPF) is proposed. Membrane-inspired algorithms can reach an evolutionary behavior based on biochemical processes to find the best parameters for generating a feasible and safe path. The proposed MemPBPF algorithm uses a combination of the structure and rules of membrane computing. In that sense, the proposed MemPBPF algorithm contains dynamic membranes that include a pseudo-bacterial genetic algorithm for evolving the required parameters in the artificial potential field method. This hybridization between membrane computing, the pseudo-bacterial genetic algorithm, and the artificial potential field method provides an outperforming path planning algorithm for autonomous mobile robots. Computer simulation results demonstrate the effectiveness of the proposed MemPBPF algorithm in terms of path length considering collision avoidance and smoothness. Comparisons with two different versions employing a different number of elementary membranes and with other artificial potential field based algorithms are presented. The proposed MemPBPF algorithm yields improved performance in terms of time execution by using a parallel implementation on a multi-core computer. Therefore, the MemPBPF algorithm achieves high performance yielding competitive results for autonomous mobile robot navigation in complex and real scenarios.

Highlights

  • Autonomous mobile robots are automatic systems with locomotion ability [1]; the level of autonomy can be determined using a rank

  • EXPERIMENTAL RESULTS AND ANALYSIS we describe the test specifications and we present the path planning results obtained with the MemPBPF algorithm considering static and dynamic environments

  • A hybrid algorithm based on membrane pseudo-bacterial potential field (MemPBPF) was proposed to solve path planning problems efficiently for autonomous mobile robot (AMR) navigation

Read more

Summary

INTRODUCTION

Autonomous mobile robots are automatic systems with locomotion ability [1]; the level of autonomy can be determined using a rank. Path planning can be applied only when the environment map is known; AMRs should be capable of performing simultaneous localization and mapping (SLAM) [5]. Both tasks can become a bottleneck, and the reduction of computational or time complexities is very important. This proposal named MemPBPF contributes to state-ofthe-art with a highly efficient path planning method for AMRs with the following three characteristics: 1) It reduces time complexity by integrating membrane computing [6], the pseudo-bacterial genetic algorithm [7], and the artificial potential field method [8].

FUNDAMENTALS
ARTIFICIAL POTENTIAL FIELD METHOD
MEMBRANE COMPUTING
EXPERIMENTAL RESULTS AND ANALYSIS
ROBOTIC PLATFORM IMPLEMENTATION
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call