Abstract
This chapter illustrates the usefulness of planning techniques—namely, HTNs and Markov Decision Processes (MDPs), for the design of a high-level navigation controller for a mobile robot. The approach is by no means limited to navigation tasks. It can be pursued for a wide variety of robotics tasks, such as object manipulation and cleaning. However, to keep the presentation technically grounded with enough details about the sensory-motor functions controlled, it is necessary to illustrate a specific task. Several sensory-motor functions are presented and discussed in the chapter. An approach that exemplifies the use of planning techniques for synthesizing alternative plans and policies for a navigation task is described. Before that, the chapter introduces the reader to the important and mature area of path and motion planning.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have