This paper presents a novel randomized motion planner for robots that must achieve a specified goal under kinematic and/or dynamic motion constraints while avoiding collision with moving obstacles with known trajectories. The planner encodes the motion constraints on the robot with a control system and samples the robot's state ¤ time space by picking control inputs at random and integrating its equations of motion. The result is a probabilistic roadmap of sampled state ¤ time points, called milestones, connected by short admissible trajectories. The planner does not precompute the roadmap; instead, for each planning query, it generates a new roadmap to connect an initial and a goal state ¤ time point. The paper presents a detailed analysis of the planner's convergence rate. It shows that, if the state ¤ time space satisfies a geometric property called expansiveness, then a slightly idealized version of our implemented planner is guaranteed to find a trajectory when one exists, with probability quickly converging to 1, as the number of of milestones increases. Our planner was tested extensively not only in simulated environments, but also on a real robot. In the latter case, a vision module estimates obstacle motions just before planning starts. The planner is then allocated a small, fixed amount of time to compute a trajectory. If a change in the expected motion of the obstacles is detected while the robot executes the planned trajectory, the planner recomputes a trajectory on the fly. Experiments on the real robot led to several extensions of the planner in order to deal with time delays and uncertainties that are inherent to an integrated robotic system interacting with the physical world.

IntroductionThis paper proposes a novel approach to As the use of Remotely Operated modeling the four quadrant dynamic response Underwater Vehicles becomes more of thrusters as used for the motion control of widespread and their tasking more complex ROV and AUV underwater vehicles. The in deeper waters, there is a need to free the significance is that these vehicles are small in vehicle from the power and signal tether, and size and respond quickly to commands.to increase both the level of control autonomy Precision in motion control will require and the maneuvering precision of these further understanding of thruster performance underwater robots. In a recent paper, Yoerger than is currently available. The model et. al. (1990) point out that Underwater includes a four quadrant mapping of the Vehicle thrusters must be properly modeled if propeller blades lift and drag forces and is good results are to be obtained for the vehicle coupled with motor and fluid system motion control. Thrusters are comprised of dynamics. A series of experiments is propellers driven by a motor -the usual way described for both long and short period in which ships have been propelled through triangular, as well as square wave inputs.the seaway since the days of commercial The model is compared favorably with sailing ships. However, while there is a long experimental data for a variety of differing history of theoretical research, experimental conditions and predicts that force overshoots validation and practical experiential are observed under conditions of rapid knowledge concerning the performance of command changes. Use of the model will ships propellers, the issues relating to the improve the control of dynamic thrust on control of Remotely Operated Vehicles these vehicles.(ROVs) and Autonomous Underwater Yoerger et. al. [I], developed a lumped parameter model of the dynamic response of an ROV thruster that went beyond the popular notion that, for a given unit with fixed pitch blading, thrust and input torque are related to the modified square of the propeller rotational rate and the angle of advance.They introduced the idea that fluid momentum considerations in the thruster shrouding area gives rise to a time lag in the response of thrust to stepwise inputs of motor torque. Experimental results under steady state conditions for single quadrant operation certainly verified the well known square law relationship between thrust and propeller rotational speed, and it did appear that the thrust response had long lag times at low thrust levels. However, little details were provided of the actual experimental thrust data under varied experimental conditions. For instance, dynamic energy balance arguments were applied but dynamic momentum arguments were ignored in the formulation of the thrust equation. Also, an instantaneous relationship between propeller rotational rate and the lumped parameter measure of flowrate was used which cannot be supported in reality.We believe that such a model is still insufficient to understand th...

Abstract. This paper studies non-gaited, multi-step motion planning, to enable limbed robots to free-climb vertical rock. The application of a multi-step planner to a real free-climbing robot is described. This planner processes each of the many underlying one-step motion queries using an incremental, sample-based technique. However, experimental results point toward a better approach, incorporating the ability to detect when one-step motions are infeasible (i.e., to prove disconnection). Current work on a general method for doing this, based on recent advances in computational real algebra, is also presented.

Thia paper describes B relative position sensing strategy that fuses monocular vlsion (a bearing measurement) with accelerometer and rate gym measurements to generate an estimate of relative position between a free-Eoatiog underwater vehicle and P ststlonary object of interest Tbh type of position estimate is a com reqoirement for interventioncapable antonomous underwater vehicles. These vehicles can perform autonomous manipulation tasks, during which the vehicle needs to control Its posltion relative to objects In Its emlroumenl. For frn8oating underwater vehicles, camera motion is generally unknown and must be e6 timated together with relative position. Various vision-only systems have been used to estimate relative position and camera motion, but these are difficult to Implement in real underwater environments.The system we propose relies on vision to generate relative position information. but also furer inerdal rate sensors to reduce the amount of information that nefds to be extracted from the vision system. The result is a system tbnt potentially is simpler and more robust than a vision-only solution. However, the use of inertial rate renson lntmdvcer sweral issues. The rate measurements are subject to biases, wbich need to be esflmated to prevent the accumulation of unbounded drill when the measurements are integrated. This problem & non-linear, which presents several challenges in the estimator design. Finally, sufficient camera motion h required for the estimator to "verge, which necessitates the design of a suitable trajectory.This paper discusses some of the implementation challenges, outlines an estimation algorithm that is uniquely adopted for this Sensor fusion problem, develops a method to generate useful vehicle trajectories, and presents Dome results from laboratory experiments with a testbed manip ulaior system. For these experiments, the estimator wm implemented U pari of a closed-loop control system that CPU perform an abject pick-up task.