Robot Path Planning Based on Genetic Algorithm Fused with Continuous Bezier Optimization. (25th February 2020)

• Record Type: Journal Article
• Authors: Ma, Jianwei; Liu, Yang; Zang, Shaofei; Wang, Lin
• Other Names: Köker Raşit Academic Editor.
• Is Part Of: Computational intelligence and neuroscience. Volume 2020(2020)
• Publication Date: 2020-02-25
• Subjects: Neurosciences -- Data processing -- Periodicals; Computational intelligence -- Periodicals; Computational neuroscience -- Periodicals; 612.80285
• DOI: 10.1155/2020/9813040
• Languages: English
• ISSNs: 1687-5265

Abstract

Abstract : In this study, a new method of smooth path planning is proposed based on Bezier curves and is applied to solve the problem of redundant nodes and peak inflection points in the path planning process of traditional algorithms. First, genetic operations are used to obtain the control points of the Bezier curve. Second, a shorter path is selected by an optimization criterion that the length of the Bezier curve is determined by the control points. Finally, a safe distance and adaptive penalty factor are introduced into the fitness function to ensure the safety of the walking process of the robot. Numerous experiments are implemented in two different environments and compared with the existing methods. It is proved that the proposed method is more effective to generate a shorter, smoother, and safer path compared with traditional approaches.