Dissertation > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Robotics > Robot

Research on the Complete Coverage Path Planning Algorithm of Mobile Robot

Author ZhaoHuiNan
Tutor LiuShuHua
School Northeast Normal University
Course Applied Computer Technology
Keywords Complete coverage path planning boustrophedon binary search robot
Type Master's thesis
Year 2011
Downloads 27
Quotes 0
Download Dissertation

In current society, we can see that the science and technology is rapidly developing now. With the development in family, business, semi-automatic and automatic service, people now could relief from their heavy cleaning work. Therefore, the cleaning robot has good market prospects, which caused great attention already in recent years; thereby complete coverage path planning technology has become a new research direction of service robot. There is a need of cover algorithms in the applications of robot use, such as mines detecting, the ground sweeping, and maps creating. In these applications, it requires the robot (or robot detector) to cover all areas that are not occupied by obstacles in that environment. The study of complete coverage technique is regarding the research on applications in mobile robot, which with a high value of study and application.For the methods of comparison of complete coverage recently, we use a modeling method by applying with grid map in this paper, because of the grid map is easier to create and maintain; the ideal coverage process is to search a route, which is unduplicated and also goes through all grid path. According to knowledge of the environment, covering algorithm can be divided into known and unknown environment covering algorithm. Under the known environment, it requires the robot to map out a path with the least cost, and can go through all the points in the environment. At this point, the problem is similar to the traveling salesman problem (TSP), so it is a NP difficulty. Under the unknown environment, it requires the robot must use its own sensors to know the environment and to draw up a coverage path plan, the task is known as sensor-based coverage job.Under the grid environment, the problem of existence of arbitrary-shape static objects or obstacles, we put forward a combination of binary search method and boustrophedon full coverage path planning algorithm, which can speed up finding the initial position in next space that is not covered; therefore, this algorithm can improve the coverage efficiency.In order to verify the feasibility of the method, we use Java programming language and emulate a variety of indoor environments to test this algorithm; the simulation results show that the algorithm is feasible indeed. In addition, by comparing with other full coverage algorithms, the results show that the proposed algorithm can effectively reduce the duplication of coverage, and it will open a better market and be widely used.

Related Dissertations
More Dissertations