Exploration, military, and humanitarian efforts call for robots capable of navigating extreme terrain. Lacking the combination of computational efficiency and understanding of robot/terrain dynamics, existing algorithms are unable to answer this call. Future robots that reliably and rapidly move over extreme terrain will generate new scientific discoveries, expand the bounds of field operations and secure a safer world. Existing navigation algorithms do not provide this capability; most lack the ability to reason about the robot/terrain interaction with sufficient fidelity. Algorithms capable of reasoning with sufficient fidelity operate too slowly and require more information about the environment than is generally available. Probabilistic planning approaches address some of the computational problems associated with high degree of freedom manipulators, but are not yet applicable to the domain of mobile robots operating with incomplete knowledge. This work explores randomized planning in conjunction with physical modeling to provide real time, sufficient fidelity plans for traversing extreme terrain. Randomized planning is introduced to address both the real time and high dimensionality aspects of the problem. Physical modeling

This paper describes an action planning method for field robots. The method represents all possible activities of the robot with discrete actions, or action modules. The action modules are assembled into a sequence, or action plan. A successful action plan is one that allows the robot to complete the task without violating any of the physical constraints of the robot or the task. These constraints are verified using physics-based models of the robot and environment. Plans are developed that explicitly consider constraints such as power, actuator saturation, wheel slip, and vehicle stability. The methodology is described in the context of planetary exploration similar to the recent NASA Mars Pathfinder mission. Theoretical analysis, simulation studies, and a laboratory demonstration of the method are presented.

Articulated Wheeled Robotic (AWR) locomotion systems consist of chassis connected to a set of wheels through articulated linkages. Such articulated “leg-wheel systems ” facilitate reconfigurability that has significant applications in many arenas, but also engender constraints that make the design, analysis and control difficult. We will study this class of systems in the context of design, analysis and control of a novel planar reconfigurable omnidirectional wheeled mobile platform. We first extend a twist based modeling approach to this class of AWRs. Our systematic symbolic implementation allows for rapid formulation of kinematic models for the general class of AWR. Two kinematic control schemes are developed which coordinate the motion of the articulated legs and wheels and resolve redundancy. Simulation results are presented to validate the control algorithm that can move the robot from one configuration to another while following a reference path. The development of two generations of prototypes is also presented briefly.

Summary. Evaluation and comparison of locomotion performance of rovers is a difficult, though very important issue. The performance is influenced by a large number of parameters. In this work, three different rovers were analyzed from a kinematic point of view. Based on a kinematic model, the optimal velocities at the actual position were calculated for all wheels and used for characterization of the suspension of the different rovers. Simulation results show significant differences between the rovers and thus, the utility of the chosen metric. It is shown that a substantial reduction of slip can be achieved by integrating kinematics in a modelbased velocity controller. 1