A Model-Based Approach to the Autonomic Management of Mobile Robot Resources

Mobile robots are increasingly used in applications as diverse as space exploration, rescue missions, automatic floor cleaning in buildings and factories, and mobile surveillance systems. However, one major challenge encountered in the development of applications involving mobile robots is the limited amount of resources that these devices have at their disposal. Effective use of mobile robot resources such as bandwidth is essential for the success of these applications. This paper presents a model-based, utility-driven approach to the autonomic management of mobile robot resources. Starting from a model of the robot resources and of the components that use these resources, the approach builds an autonomic solution capable of optimising a utility function supplied by the robot administrator. This self-optimisation involves adapting the operation mode of the robot components to changes in the amount of resources available and in the environment. The approach is validated through a case study that involves the self-optimising management of bandwidth for a Pioneer 2AT-8 mobile robot. Keywords-model-based self-X adaptation, mobile robot, utility function, resource management