Intelligent Mobile Robot Map Description And Navigation

Environment modeling and autonomous navigation are two important issues for robot. Intelligent mobile robots have begun to be made use of in all kinds of fields. Such as explore the unknown environment, comprehensive rescue, military reconnaissance, object distribution, mapping and measuring as well as family services, and with progress of the intelligence level of robot and improving of manufacturing, I believe the intelligent mobile robots will be more popular what ever for civil use or in military field.Environment modeling is the first issue for Mobile robot to achieve various functions. Environment is usually divided into three types[1]:unknown environment,known environment and partly known environment.The environment is usually including spatial information, however, with the development of high autonomy and intelligent robots, deeper knowledge acquisition becomes possible, and this knowledge will be more conducive to the robot operation (navigation and positioning), This article will focus on the semantic map model of knowledge representation, which can help the robot to do more complex mission planning. First, define a specific type of semantic mapping, then represent it with the topology model, and access to environmental information using a variety of sensors, through the feature extraction, semantic description of the Environment, then use the builted environment model for robot to achieve navigation task.The goal of navigation is for a robot to find a path from the starting point to target point, the robot in the process of motion can avoid all the obstacles encountered, and optimized to meet certain targets as much as possible such as shortest time or lowest energy consumption and so on. Mobile robot with maps and navigation on the various issues described in this paper mainly been studied in the following aspects:Firstly discuss the representation of a map. Discuss how to describe the environment with the semantic map, to achieve a formal description with the topological graph for the semantic information, define the meaning of the topological nodes and edges, and define how the topology map is stored.Followed discuss a radar data processing algorithms, the data obtained by cluster analysis of laser radar, to study how to extract simple semantic information from laser radar sensors, such as wall, corner, road characteristics, intersection attributes.Finally, according to the environmental modeling and feature extraction of laser radar discuss the design of navigation algorithm. study how to use laser radar to achieve barrier avoidance.We proposed a navigation algorithm based on electronic compass and GPS, working around the problems of vibration control and junctions error identification which due to the GPS positioning error, and discussed how to use the extended Kalman filter to improve positioning accuracy, and presents a Fast combination of location algorithm, discussed a algorithm which combine GPS and laser radar to recognize junctions.