The primary focus is on providing a formal basis for behavior- based robotics using techniques that have been successful in control-based approaches for steering and stabilizing robots that are subject to nonholonomic constraints. In particular, behaviors for robots are formalized in terms of kinetic state machines, a motion description language and the interaction of the kinetic state machine with information coming in from (limited range) sensors. This allows us to create a mathematical basis for discussing these systems, including techniques for integrating sets of behaviors. In addition we suggest optimality criteria for comparing both atomic and compound behaviors in various environments. A hybrid architecture for the implementation of path planners that use the motion description language is presented. The design and implementation of a planner for path planning and examples of obstacle avoidance with nonholonomic robots are discussed.