A distributed algorithm for mobile robot localization and mapping in wireless sensor networks

This paper presents a navigation and mapping technique that integrates the global view of a wireless sensor network and the local advance sensing of mobile robots. Search and rescue missions rely on robots that can successfully navigate through an indoor environment which may be undefined and filled with unpredictable obstacles. The onboard sensors and local navigation schemes can provide obstacle avoidance and help robots navigate locally, but not for a long distant goal due to lack of global knowledge and problems of local minima. Therefore some help from the environment were proposed in many previous works such as visual landmarks, or radio signal transmitted from Wi-Fi beacons or wireless sensor networks. The latter, which involves many sensor nodes, if deployed in well-organized formation, can serve as signposts that will help mobile robots for navigation. Hence the need for a predefined map can be eliminated. By utilizing robots on-board sensors to process the received data from near-by nodes and combining it with the on-board sensed data, the algorithm we propose will provide the ad hoc SLAM capability to both robots and wireless sensor network which in turn can guide the robot to its destination at the same time.