In this paper, a novel approach to automated inspection is presented, which uses a mobile robot equipped with a 2D laser rangefinder. The main idea underlying the proposed method is that of comparing current laser readings with local range data of the environment stored in a database, to look for new or removed objects. First, the robot is guided to reach goals, fixed in critical areas where inspection is required. Range data of the region surrounding each goal are automatically acquired and stored in a database. Afterwards, the robot can begin its surveillance task. Each time it reaches a goal, comparison between the current readings and the stored local data is performed using an Iterative Closest Point (ICP)-based scan-matching algorithm. Then, a fuzzy logic inference system establishes whether a significant variation of the scene has occurred and an alarm signal must be produced. Experimental results show that the proposed approach is reliable in detecting either new or missing objects and can be effectively used in automated surveillance systems in dynamic environments.

This content is only available via PDF.
You do not currently have access to this content.