Escaping route method for a trap situation in local path planning

Abstract

This paper introduces a new framework for escaping from a local minimum in path planning based on artificial potential functions (APFs). In particular, this paper presents a set of analytical guidelines for designing potential functions to avoid local minima in a trap situation (in this case, the robot is trapped in a local minimum by the potential of obstacles). The virtual escaping route method is proposed to allow a robot to escape from a local minimum in a trap situation where the total forces are composed of repulsive forces by obstacles and attractive force by a goal are zero. The example results show that the proposed scheme can effectively construct a path planning system with the capability of reaching a goal and avoiding obstacles, despite a trapped situation under possible local minima.

Vector Field UAV Guidance for Path Following and Obstacle Avoidance with Minimal Deviation

“…PF local minimums occur when closely spaced obstacle potentials produce a well on the descent gradient where a premature stable point is reached. Proposed solutions to local minimum include object clustering and virtual waypoint method [35], virtual escaping route [36], and use of navigation functions [22]. Oscillations in PF were addressed in [37,38].…”

Improving Lateral Safety Distance-Based on Feature Detection and Probabilistic Roadmaps for Unmanned Vehicle Path Planning

“…It is designed to enable an autonomous vehicle to achieve a series of collision-free and safety motions in a given environment to accomplish certain tasks. Traditional path searching and planning algorithms have many types, such as rapidly exploring random tree path planning, sub-goal networks, A* algorithms and D* algorithms, artificial potential fields, particle swarm optimization (PSO) algorithms, and polynomial interpolation methods [1][2][3][4]. With the increasing complexity of the environment and the difficulty of tasks, these traditional methods are, however, often incapable of achieving the ultimate performance.…”

Navigational Control Analysis of Two-Wheeled Self-Balancing Robot in an Unknown Terrain Using Back-Propagation Neural Network Integrated Modified DAYANI Approach

“…9 Therefore, various heuristic methods have been developed to overcome the inefficiency of the classical methods. 10 Kim 11 has proposed a virtual escaping route method based on artificial potential functions for avoiding a trap situation in local minima. Ismail et al 12 have developed an idea for reducing the number of steps involved for robot motion from start to goal point by using genetic algorithm in a static grid environment.…”