Early Detection of Obstacle to Optimize the Robot Path Planning

Abstract

Robot path planning is one of the core issues in robotics and its application. Optimizing the route discovery becomes more important while dealing with the robot-based application. This paper proposes the concept of early detection of the obstacle present in the workspace of the robots. To early detect the obstacle, this paper proposes the concept of a snake algorithm along with the traditional path planning algorithms. The contour detection part is merged with the different path planning algorithms to optimize the robot traversing and benefit it in producing good results. Obstacle-free optimized path is one of the core requirements for robots in any application. With the help of path planning algorithms, robots are enabled to derive those paths in a specific environment. The presence of an obstacle makes it difficult for any path planning algorithms to derive a smooth path. The purpose of using the snake algorithm is to detect an obstacle early. This method not only perceives the obstacle but also catches out the complete boundary of the obstacle, it, thus, provides the details of obstacle coordinates to the path planning algorithm. Conceiving the complete periphery of obstacles can have multiple advantages in many application areas. A*, PRM, RRT, and RRT Smooth algorithms are considered along with the snake algorithm to validate our work in three different experimental scenarios: Maze, Random Obstacles, and Dense case. Path length, Time-taken, and Move count are parameters taken to observe the results. The result obtained using the snake algorithm with four path planning algorithms is analyzed and compared in detail with the core A*, PRM, RRT, and RRTS. Finally, the result obtained using the proposed methodology gives some encouraging results and also predicts the exploration of the robot’s path planning for more applications and fields.