Development of path planning algorithm for biped robot using combined multi-point RRT and visibility graph

• Published in: International journal of information technology (Singapore. Online) Vol. 13; no. 4; pp. 1513 - 1519
• Main Authors: Pradhan, Sarthak, Mandava, Ravi Kumar, Vundavilli, Pandu R.
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01.08.2021
• Subjects: Artificial Intelligence; Computer Imaging; Computer Science; Image Processing and Computer Vision; Machine Learning; Original Research; Pattern Recognition and Graphics; Software Engineering; Vision; Path planning; Biped robot; Rapidly exploring random trees; Visibility graphs
• ISSN: 2511-2104; 2511-2112
• DOI: 10.1007/s41870-021-00696-w

Path planning plays an important role while moving the robot from one location to another. The present paper is concerned with solving the path planning problem of the biped robot in a static environment. Here, an attempt is made to combine the benefits of established rapidly exploring random trees (RRT) algorithm and visibility graph algorithm. Various approaches, such as normal RRT that starts the search from start and goal points, multi-node point RRT approach in which a random node is generated somewhere in between the start and goal points, and a tree is started from each of these points as the root, has been developed. Further, visibility graph has also been applied to smooth or optimize the path obtained in the previous step. The developed path planning algorithms are initially tested in computer simulations. For experimental purposes, a global camera attached at the top of the terrain is used to map the environment. Further, a humanoid with 18 degrees of freedom is used to execute the path decided by the algorithm. Required corrections are made at regular intervals if the robot deviates from its path.