Constraint-Based Task Selection and Configuration for Autonomous Mobile Robots

An autonomous mobile robot must be capable of rationally selecting tasks to perform. A task is any robotic action or series of actions to achieve some state or goal. Rule-based systems provide an intuitive means of encoding system knowledge into a set of rules that guide the robot. The dynamic nature of the world is seldom consistent enough to support the rigidity of rules. A decision maker is needed that provides the expressability of the problem model as rule-based systems, but provide flexibility, extensibility and generalizability. The constraintbased methods can be used to model the problem of selecting and configuring tasks for mobile robots. Constraint Satisfaction Problems (CSPs) will provide a framework in which multiple conflicting constraints upon the robot can be resolved in such a way so that the robot will not only perform correctly, but also will meet or exceed its performance requirements. Constraints also provide an intuitive means of specifying the decision model, but without the rigidity of rules. Performance objectives are incorporated into the constraint model so thatthe decision making system will be capable of rationally guiding the robot through actions that best meet current needs and goals. A framework and decision maker will be presented for which robot tasks are modeled and autonomously selected. The framework is demonstrated using simulation experiments for a delivery robot, a polar mobile robot, and a robot for autonomous urban search and rescue.