Framework transformation for local information on artificial potential field path planning

This paper addresses the issue of local information on the Artificial Potential Field (APF). The APF is designed as the path planning with global information usually. For the real time platform in the vast area, it is hard for the sensor to acquire data environment in the global view method. Generally, the sensor is attached on the robot itself that only could be obtained the local information. This paper proposes an approach to handle the local information on the APF. Framework transformation is the key idea to cope the problem. With the integration of the image processing, Voronoi Diagram, and framework transformation, the initial, goal, and obstacles from the real world coordinate can be determined in the APF environment scenario. The transformation of two-dimensional image is used to generate the APF. The global optimum in the local information is as waypoints for the global optimum in whole environment scenario. Local data set was used to test the performance of the algorithm. Two scenarios were used in this research, i.e. the static environment and dynamic environment with moving obstacle. The moving obstacle was moved toward to the robot. The results show that the concept can be used for APF that uses local information.