A time-sensitive approach to mobile robot autonomous navigation

The goal of autonomous mobile robotics is to build physical systems that can interact with environments not specifically structured for them. A mobile robot to move autonomously inside an environment must be able to self-localise. When the robot can not exploit a model of the environment to self-localise, it should be able both to self-localise and to build a representation of the workspace it is exploring. This is a hard issue since localisation needs an environment representation, and the construction of an environment representation needs the robot position. Which came first? The approach we propose answers the question asserting that the two activities may be executed concurrently since they act on different time scales. The approach, titled CLAM (Concurrent Localisation And Mapping), is based on the conjecture that Localisation and Modelling, acting on different time scales, are mostly independent each other. When a synchronisation is needed, it is controlled by an external and suitable strategy. A concrete system based on CLAM principles has been developed exploiting a software architecture for time sensitive systems.