Bezier Curve Collision-Free Route Planning Using Meta-Heuristic Optimization

Abstract


 
 
 
 A collision-free route is very important for achieving sustainability in a manufacturing process and vehicle robot trajectories that commonly operate in a hazardous environment surrounded by obstacles. This paper presents a collision avoidance algorithm using a Bezier curve as a route path. The route planning is modeled as an optimization problem with the objective optimization is to minimize the route length considering an avoiding collision constraint. The collision-avoidance algorithm based on curve point analysis is developed incorporating metaheuristic optimizations, namely a Genetic Algorithm (GA) and a Grey Wolf Optimizer (GWO). In the collision avoidance algorithm, checking of curve point's position is important to evaluate the individual fitness value. The curve points are analyzed in such a way so that only the paths which are outside the obstacle area are selected. In this case, besides the minimum length as a fitness function, the constraint is the position of curve points from an obstacle. With the help of meta-heuristic optimization, the developed collision avoidance algorithm has been applied successfully to different types of obstacle geometries. The optimization problem is converted to the maximization problem so that the highest fitness value is used to measure the performance of the GA and GWO. In general, results show that the GWO outperforms the GA, where it exhibits the highest fitness value. However, the GA has shown better performance for the narrow passage problem than that of the GWO. Thus, for future research, implementing the hybrid technique combining the GA and the GWO to solve the advanced path planning is essential.