ABSTRACT
There is a demand for autonomous mobile robots in various fi elds of application, such as material transport, cleaning, monitoring, guiding people and military applications. These mobile robots must interact with their environment to accomplish their tasks, this environment may be subject to change environments and unforeseen. This chapter addresses the problem
of navigation of a mobile robot, understanding navigation as the move from point A (start position) to point B (target position) through a defi ned path. The mobile robot will have to interact with the environment and avoid the obstacles, following the path planned to achieve its established mission. All of these tasks, without the assistance of a human operator. To achieve this, we propose the integration of the method of artifi cial potential fi eld (APF) with a genetic algorithm and parallel computing to develop a simulation of high-performance navigation system. This works is organized as follows: Section 12.2 shows the fundamentals of the artifi cial potential fi eld, which is a mathematical method widely used in the path planning for mobile robots, given that it provides an effective control on the movement (Kim et al. 2011). Section 12.3 shows a little reminder of the genetic algorithms. In this work a genetic algorithm is used to mitigate the limitations presented by the artifi cial potential fi eld method. Section 12.4 presents the implementation of the high-performance navigation system. This section is divided into three phases of development. The fi rst phase shows the implementation of a simple navigation system where only the method of artifi cial potential fi eld is used. The second phase incorporates a genetic algorithm to achieve the implementation of the complete navigation system, which it operates with complete autonomy. Finally the third phase of the implementation the high-performance navigation system is made. Making use of the high performance parallel computation to computing proportional gains and as a consequence of this, maximizing the performance already achieved faster. In Section 12.5, the conclusions and results of this chapter are presented, the section shows the performance results of the implementations made in phases 2 and 3 of Section 12.4 in terms of performance time and the overall conclusions of this chapter.
