Showing papers on "Obstacle avoidance published in 1996"

The curvature-velocity method for local obstacle avoidance

Abstract: We present a new method for local obstacle avoidance by indoor mobile robots that formulates the problem as one of constrained optimization in velocity space. Constraints that stem from physical limitations (velocities and accelerations) and the environment (the configuration of obstacles) are placed on the translational and rotational velocities of the robot. The robot chooses velocity commands that satisfy all the constraints and maximize an objective function that trades off speed, safety and goal-directedness. An efficient, real-time implementation of the method has been extensively tested, demonstrating reliable, smooth and speedy navigation in office environments. The obstacle avoidance method is used as the basis of more sophisticated navigation behaviors, ranging from simple wandering to map-based navigation.

Sequential composition of dynamically dexterous robot behaviors

Abstract: In this thesis we present a technique for the composition of robot control laws in dynamical environments. We propose a challenging robotic task, called Dynamical Pick and Place, in which a robot equipped with merely a soft paddle must capture and contain a ball, safely negotiate it past obstacles, and bring it to rest at a desired location. We develop a composition technique for local controllers that provides a formal guarantee of the stability of the switching behavior required in this task, and provide descriptive statistics of a working implementation. Our robotic system displays unusually dexterous behavior in the face of significant system noise, and recovers gracefully from large unexpected perturbations caused by the experimenters. Our approach to controller composition makes use of the funnel as a metaphor for asymptotic stability, is motivated by the pre-image backchaining techniques developed by Lozano-Perez, Mason and Taylor, and extends their ideas from quasi-static environments to systems with full dynamics. We introduce the concepts of "dynamical obstacle avoidance" and "dynamical safety" for systems with only intermittent control of their environment, and show that it is important not only that the system avoid obstacles directly, but also that the system will never reach an obstacle before getting another chance to effect control. The Dynamical Pick and Place problem addressed by this thesis is a difficult control problem, but an easy planning problem. The system we develop provides a way to engage more powerful AI planning tools without sacrificing access to the stability arguments of dynamical systems theory.

An on-line method to evolve behavior and to control a miniature robot in real time with genetic programming

Abstract: We present a novel evolutionary approach to robotic control of a real robot based on genetic programming (GP). Our approach uses GP techniques that manipulate machine code to evolve control programs for robots. This variant of GP has several advantages over a conventional GP system, such as higher speed, lower memory requirements, and better real-time properties. Previous attempts to apply GP in robotics use simulations to evaluate control programs and have difficulties with learning tasks involving a real robot. We present an on-line control method that is evaluated in two different physical environments and applied to two tasks—obstacle avoidance and object following—using the Khepera robot platform. The results show fast learning and good generalization.

Inverse kinematics of discretely actuated hyper-redundant manipulators using workspace densities

Abstract: Hyper-redundant manipulators present an alternative to conventional 6 DOF manipulators for inspection, space, and medical applications. The additional degrees of freedom facilitate obstacle avoidance and allow tasks to be performed even if some of the actuators fail. In this paper the authors consider hyper-redundant manipulators that are actuated discretely, e.g. using two-state actuators or motors with finite resolution. The inverse kinematics problem for a discretely actuated manipulator is intrinsically different from the one for its continuously actuated counterpart. The authors present a framework for the discussion of the discretely actuated case and propose an algorithm for the inverse kinematics. The algorithm generates solutions in linear time with respect to the number of manipulator actuators, as opposed to the exponential time required by brute force search.

Experimental robotics IV : the 4th International Symposium, Stanford, California, June 30-July 2, 1995

Abstract: Collective and cooperative group behaviours: Biologically inspired experiments in robotics.- Distributed robotic manipulation: Experiments in minimalism.- A general framework for multi-robot cooperation and its implementation on a set of three hilare robots.- Cooperative autonomous low-cost robots for exploring unknown environments.- An object-oriented framework for event-driven dextrous manipulation.- Toward obstacle avoidance in intermittent dynamical environments.- Integrating grasp planning and visual servoing for automatic grasping.- contact and grasp robustness measures: Analysis and experiments.- Performance limits and stiffness control of multifingered hands.- Real-time vision plus remote-brained design opens a new world for experimental robotics.- Experimental validation of an active visual control scheme based on a reduced set of image parameters.- Task oriented model-driven visually servoed agents.- Experiments in hand-eye coordination using active vision.- Visual positioning and docking of non-holonomic vehicles.- An intelligent observer.- The development of a robotic endoscope.- The extender technology: An example of human-machine interaction via the transfer of power and information signals.- Coordinated and force-feedback control of hydraulic excavators.- Experiments with a real-time structure-from-motion system.- Robotic perception of material: Experiments with shape-invariant acoustic measures of material type.- Multi-level 3D-tracking of objects integrating velocity estimation based on optical flow and kalman-filtering.- Experimental approach on artificial active antenna.- Low cost sensor based obstacle detection and description.- Parameter sensitivity analysis for design and control of tendon transmissions.- Stiffness isn't everything.- In pursuit of dynamic range: Using parallel coupled actuators to overcome hardware limitations.- Total least squares in robot calibration.- Symbolic modelling and experimental determination of physical parameters for complex elastic manipulators.- Learning compliant motions by task-demonstration in virtual environments.- Motion control for a hitting task: A learning approach to inverse mapping.- Experimental verification of progressive learning control for high-speed direct-drive robots with structure flexibility and non-collocated sensors.- Accurate positioning of devices with nonlinear friction using fuzzy logic pulse controller.- Platooning for small public urban vehicles.- Robust vehicle navigation.- Dynamic analysis of off-road vehicles.- An autonomous guided vehicle for cargo handling applications.- Robots that take advice.- Towards principled experimental study of autonomous mobile robots.- Mission programming: Application to underwater robots.- Specification, formal verification and implementation of tasks and missions for an autonomous vehicle.- Experimental study on modeling and control of flexible manipulators using virtual joint model.- Experimental research on impact dynamics of spaceborne manipulator systems.- An operational space formulation for a free-flying, multi-arm space robot.- Experimental research of a nonholonomic manipulator.- Mobile manipulation of a fragile object.- Empirical verification of fine-motion planning theories.- Estimating throughput for a flexible part feeder.- Interest of the dual hybrid control scheme for teleoperation with time delays for proceeding of ISER'95.- Robot force control experiments with an actively damped compliant end effector.- Improved force control for conventional arms using wrist-based torque feedback.- Indoor navigation of an inverse pendulum type autonomous mobile robot with adaptive stabilization control system.- Motion and perception strategies for outdoor mobile robot navigation in unknown environments.- Programming symmetric platonic beast robots.- An experimental study on motion control of a biped locomotion machine using reaction wheels.- Real-time programming of mobile robot actions using advanced control techniques.

A hybrid GP/GA approach for co-evolving controllers and robot bodies to achieve fitness-specified tasks

Abstract: Evolutionary approaches have been advocated to automate robot design. Some research work has shown the success of evolving controllers for the robots by genetic approaches. As we can observe, however, not only the controller but also the robot body itself can affect the behavior of the robot in a robot system. We develop a hybrid GP/GA approach to evolve both controllers and robot bodies to achieve behavior-specified tasks. In order to assess the performance of the developed approach, it is used to evolve a simulated agent, with its own controller and body, to do obstacle avoidance in the simulated environment. Experimental results show the promise of this work. In addition, the importance of co-evolving controllers and robot bodies is analyzed and discussed.

Characteristics of voluntary visual sampling of the environment for safe locomotion over different terrains.

Abstract: The characteristics of visual sampling required for successful locomotion over various terrains is the focus of this work. In the first experiment we directly address the role of continuous visual monitoring of the environment in guiding locomotion by allowing the subjects to choose when and where to take a visual sample of the terrain and examine the effects of different terrains on characteristics of visual sampling. Young subjects walked over travel paths of varying difficulties while wearing opaque liquid crystal eyeglasses and pressed a hand-held switch to make the glasses transparent when they needed to sample the environment. Travel time and visual sampling characteristics were recorded. Results show that intermittent sampling (less than 50%) of the environment is adequate for safe locomotion, even over a novel travel path. The frequency, duration and timing of visual samples are dependent on terrain characteristics. Visual sampling of the environment is unaffected by preview restriction of the travel path and is increased when specific foot placement is required and there is a potential hazard in the travel path. In the second experiment we dissociated steering control and obstacle avoidance from specific foot placement and examined visual sampling demands prior to the initiation of the swing phase and during the swing phase. The results show that steering control and obstacle avoidance do influence the visual sampling time, which is scaled to the magnitude of change. Vision was used in a feedforward control mode to plan for and initiate appropriate changes in the swing limb trajectory: its use during the swing phase to provide on-line control was minimal.

Visual Navigation

Abstract: Visual navigation is a major goal in machine vision research, and one of both practical and basic scientific significance. The practical interest reflects a desire to produce systems which move about the world with some degree of autonomy. The scientific interest arises from the fact that navigation seems to be one of the primary functions of vision in biological systems. Navigation has typically been approached through reconstructive techniques since a quantitative description of the environment allows well understood geometric principles to be used to determine a course. However, reconstructive vision has had limited success in extracting accurate information from real-world images. This thesis argues that a number of basic navigational operations can be realized using qualitative methods based on inexact measurement and pattern recognition techniques. Navigational capabilities form a natural hierarchy beginning with simple abilities such as orientation and obstacle avoidance, and extending to more complex ones such as target pursuit and homing. Within a system, the levels can operate more or less independently, with only occasional interaction necessary. This thesis considers three basic navigational abilities: passive navigation, obstacle avoidance, and visual homing, which together represent a solid set of elementary, navigational tools for practical applications. It is demonstrated that all three can be approached by qualitative, pattern-recognition techniques. For passive navigation, global patterns in the spherical motion field are used to robustly determine the motion parameters. For obstacle avoidance, divergence-like measurements on the motion field are used to warn of potential collisions. For visual homing an associative memory is used to construct a system which can be trained to home visually in a wide variety of natural environments. Theoretical analyses of the techniques are presented, and implementation and testing of working systems described.

A biologically inspired neural net for trajectory formation and obstacle avoidance

Abstract: In this paper we present a biologically inspired two-layered neural network for trajectory formation and obstacle avoidance. The two topographically ordered neural maps consist of analog neurons having continuous dynamics. The first layer, the sensory map, receives sensory information and builds up an activity pattern which contains the optimal solution (i.e. shortest path without collisions) for any given set of current position, target positions and obstacle positions. Targets and obstacles are allowed to move, in which case the activity pattern in the sensory map will change accordingly. The time evolution of the neural activity in the second layer, the motor map, results in a moving cluster of activity, which can be interpreted as a population vector. Through the feedforward connections between the two layers, input of the sensory map directs the movement of the cluster along the optimal path from the current position of the cluster to the target position. The smooth trajectory is the result of the intrinsic dynamics of the network only. No supervisor is required. The output of the motor map can be used for direct control of an autonomous system in a cluttered environment or for control of the actuators of a biological limb or robot manipulator. The system is able to reach a target even in the presence of an external perturbation. Computer simulations of a point robot and a multi-joint manipulator illustrate the theory.

Visual Control of Obstacle Avoidance During Locomotion: Strategies in Young Children, Young and Older Adults

Abstract: The focus of this chapter is on understanding how obstacle avoidance during locomotion is affected by normal aging process and how this adaptability in locomotor system develops as children acquire independent bipedal locomotion. Obstacle avoidance paradigms offer a rich source of material for understanding the unique sensorimotor integration common to many visually guided movements. Based on studies on young healthy adults, we have proposed a jigsaw puzzle metaphor summarizing the key ingredients for successful obstacle avoidance. The nature of visual and kinesthetic input and the contribution of the effector system properties form the pieces of the puzzle. Studies on healthy older adults reveal relatively well preserved obstacle avoidance strategies, although there are some differences when compared to the healthy young adults. Deterioration in sensory input and effector system characteristics shows up as adaptive changes in feedforward control of limb trajectory over obstacles. This suggests that the puzzle is relatively robust with cracks appearing in some pieces. Preliminary studies on children provide interesting signposts for the development of stable obstacle avoidance strategies. High failure rate and poorer control of limb trajectory over obstacles characterize the gait patterns of young children in a cluttered environment. This suggests that the pieces of the puzzle have to be sculpted and merged into a coherent picture during the development process.

Local incremental planning for a car-like robot navigating among obstacles

Abstract: We present a local approach for planning the motion of a car-like robot navigating among obstacles, suitable for sensor-based implementation. The nonholonomic nature of the robot kinematics is explicitly taken into account. The strategy is to modify the output of a generic local holonomic planner, so as to provide commands that realize the desired motion in a least-squares sense. A feedback action tends to align the vehicle with the local force field. In order to avoid the motion stops away from the desired goal, various force fields are considered and compared by simulation.

Real-time single-workstation obstacle avoidance using only wide-field flow divergence

Abstract: A real-time robot vision system is described which uses only the divergence of the optical flow field for both steering control and collision detection. The robot has wandered about the lab at 20 cm/s for as long as 26 minutes without collision. The entire system is implemented on a single ordinary UNIX workstation without the benefit of real-time operating system support. Dense optical flow data are calculated in real-time across the entire wide-angle image. The divergence of this optical flow field is calculated everywhere and used to control steering and collision behavior. Divergence alone has proven sufficient for steering past objects and detecting imminent collision. The major contribution is the demonstration of a simple, robust, minimal system that uses flow-derived measures to control steering and speed to avoid collision in real time for extended periods. Such a system can be embedded in a general, multi-level perception/control system.

Avoidability measure in moving obstacle avoidance problem and its use for robot motion planning

Abstract: This paper presents a new solution approach to moving obstacle avoidance problem of a robot. A new concept avoidability measure (AVM) is defined to describe the state of a pair of a robot and an obstacle regarding the collision between them. As an AVM, virtual distance function (VDF) is derived as a function of the distance from the obstacle to the robot and outward speed of the obstacle relative to the robot. By keeping the virtual distance above some positive limit value, the robot avoids the obstacle. In terms of the VDF, an artificial potential field is constructed to repel the robot away from the obstacle and to attract the robot toward a goal location. At every sampling time, the artificial potential field is updated and the force driving the robot is derived from the gradient of the artificial potential field. The suggested algorithm drives the robot to avoid moving obstacles in real time. Since the algorithm considers the mobility of the obstacle as well as the distance, it is effective for moving obstacle avoidance. Some simulation studies show the effectiveness of the proposed approach.

Obstacle avoidance during locomotion is unaffected in a patient with visual form agnosia.

Abstract: A patient (D.F.) who developed visual form agnosia following carbon monoxide-induced anoxia was assessed on three tests designed to measure her sensitivity to obstacle height in a locomotor task. Although her verbal estimates of the height of the obstacles were correlated with their actual height, the slope of the line relating estimated and actual obstacle height was much shallower than for control subjects. Similarly, when asked to estimate the height of the obstacle by raising one leg while standing nearby, the slope of line relating toe elevation and obstacle height was shallower than in control subjects. In contrast, D.F. was able to negotiate the same obstacles during locomotion as well as control subjects: toe elevation increased linearly as a function of obstacle height with similar slopes and correlation for the line relating toe elevation and obstacle height. These results provide additional support for the proposal by Goodale and Milner that the cortical pathways mediating the required transformations for the visual control of skilled actions are separate from those mediating experiential perception of the visual world.

Obstacle avoidance in a terrestrial mobile robot provided with a scanning retina

Abstract: This study describes the role of a novel visual sensor, called the "scanning local motion detector" (SLMD) for the visual navigation of a terrestrial mobile agent. Following a brief theoretical description of the principle of retinal scanning coupled with notion parallax, we show that a single retina of 24 pixels - when oriented in the mobile agent's forward line of move - allows the latter to avoid the visually detected obstacles in a reflex manner by correcting its trajectory in real time. The results of the preliminary computer simulations yield credit to a zig-zag algorithm which allows the agent to avoid the oncoming obstacles while immediately "keeping an eye" on its lateral regions. In parallel with the simulation, we designed and built a hardware prototype of a scanning retina with coarse optical resolution (average angular sampling of 3/spl deg/) mounted on top of a circular mobile platform. The preliminary results of the robot's successful performances confirm the validity of the zig-zag algorithm.

Evolving Obstacle Avoidance Behavior in a Robot Arm

Abstract: Existing approaches for learning to control a robot arm rely on supervised methods where correct behavior is explicitly given. It is di cult to learn to avoid obstacles using such methods, however, because examples of obstacle avoidance behavior are hard to generate. This paper presents an alternative approach that evolves neural network controllers through genetic algorithms. No input/output examples are necessary, since neuroevolution learns from a single performance measurement over the entire task of grasping an object. The approach is tested in a simulation of the OSCAR-6 robot arm which receives both visual and sensory input. Neural networks evolved to e ectively avoid obstacles at various locations to reach random target locations.

A real-time planning algorithm for obstacle avoidance of redundant robots

Abstract: A computationally efficient, obstacle avoidance algorithm for redundant robots is presented in this paper. This algorithm incorporates the neural networks and pseudodistance function Dp in the framework of resolved motion rate control. Thus, it is well suited for real-time implementation. Robot arm kinematic control is carried out by the Hopfield network. The connection weights of the network can be determined from the current value of Jacobian matrix at each sampling time, and joint velocity commands can be generated from the outputs of the network. The obstacle avoidance task is achieved by formulating the performance criterion as Dp>dmin (dmin represents the minimal distance between the redundant robot and obstacles). Its calculation is only related to some vertices which are used to model the robot and obstacles, and the computational times are nearly linear in the total number of vertices. Several simulation cases for a four-link planar manipulator are given to prove that the proposed collision-free trajectory planning scheme is efficient and practical.

Neural networks for robotic control : theory and applications

Abstract: An Overview of Neural Networks in Control Applications. Artificial Neural Network-Based Intelligent Robot Dynamic Control. Neural Servo Controller for Position, Force and Stabbing Control of Robotic Manipulators. Model-Based Adaptive Neural Structures for Robotic Systems. Intelligent Coordination of Multiple Systems with Neural Networks. Neural Networks for Mobile Robot Piloting Control. A Neural Network Controller for the Navigation and Obstacle Avoidance of a Mobile Robot. An Ultrasonic 3-D Robot Vision System Based on the Statistical Properties of Artificial Neural Networks. Brain Building for a Biological Robot. Robustness of a Distributed Neural Network Controller for Locomotion in a Hixapod Robot.

A force feedback joystick and control algorithm for wheelchair obstacle avoidance

Abstract: Many powered wheelchair users have difficulty manoeuvring in confined spaces. Common tasks such as traversing through doorways, turning around in halls or travelling on a straight path are complicated by an inability to accurately and reliably control the wheelchair with a joystick or other common input device, or by a sensory impairment that prevents the user from receiving feedback from the environment. An active joystick with force feedback to indicate obstacles in the environment has been developed. Two force feedback schemes designed to assist a powered wheelchair user have been developed and implemented using the active joystick. The development of the joystick and associated control algorithms are described.

An Evolutionary Robot Navigation System Using a Gate-Level Evolvable Hardware

Abstract: Recently there has been a great interest in the design and study of evolvable systems based on Artificial Life principles in order to control the behavior of physically embedded systems such as a mobile robot. This paper studies an evolutionary navigation system for a mobile robot using a Boolean function approach implemented on gate-level evolvable hardware (EHW). The task of the mobile robot is to reach a goal represented by a colored light while avoiding obstacles during its motion. Using the evolution principles to build the desired behaviors, we show that the Boolean function approach using gate-level evolvable hardware is sufficient. We demonstrate the effectiveness of the generalization ability of EHW by comparing the method with a Boolean function approach implemented on a random access memory (RAM). The results show that the evolvable hardware system obtains the desired behaviors in twice fast time and that the EHW generates a robust robot behavior insensitive to the robot position and the obstacles configurations.

Safeguarded Teleoperation for Lunar Rovers

Abstract: In this paper we present recent advances in developing and validating the safeguarded teleoperation approach to timedelayed remote driving. This approach shares control of the rover using a command fusion strategy: In benign situations, usus remotely drive the rover; in hazardous situations, a safeguarding system running on-board the rover overwrites user commands to ensure vehicle safety. This strategy satisfies users, because it allows them to drive (except in hazardous situations), while maintaining the integrity of the rover and mission. We present results from experiments on untrained teleoperators with and without safeguarding, which reveal needs to be met by future user interfaces. We describe three technical advances in safeguarding: improving the accuracy of dead reckoning by a factor of 2. speeding up the controller by a factor of 18, and developing an area-based rather than a path-based obstacle avoidance planner in order to circumvent map merging problems. Finally, we discuss a field trial validating the a p proach in a 10 km traverse, demonstrating the effectiveness of safeguarding, even with malicious drivers.

Estimation of Mobile Robot Localization: Geometric Approaches

Abstract: The real device is shown in Fig. 27.1. It is a cart-like wheeled vehicle sketched on Fig. 27.2. It is capable to perform planar displacements and its configuration q (Eq. 27.1) is composed of the 2-D coordinates (x c y c ) of a characteristic point together with the orientation θ defined in a world coordinate W (Fig. 27.2).

Dynamics parametrically controlled by image correlations organize robot navigation

Abstract: Behavior-based robot designs confront the problem how different elementary behaviors can be integrated. We address two aspects of this problem, the stabilisation of decisions based on changing behavioral requirements and the fusion of multiple sources of qualitative sensory information. These issues are studied in the context of a vision-guided mobile robot that is endowed with the ability to reach a goal while it avoids obstacles. Behavior is organized from the “inside” of the robot. Even in the absense of external stimuli the internal dynamics generate behavior. By exploiting image correlations the visual sensors provide coarse qualitative estimates of spatial relations. These estimates are immediately coupled into neural dynamics realized by neural fields that generate obstacle avoidance and homing behaviors of an autonomous mobile robot. The background of the neural field approach is theoretical work on the function of cerebral cortex ([1]). Neural fields represent a state space approach to estimation and control. Convergent information cooperates and divergent information competes in shaping the stable attractor states. The states of the dynamical systems are on the one hand instrumental for the control of behavior on the other hand they provide concise hypotheses for the interpretation of sensory input. The concept of dynamic state deals effectively with contradictory information and assures that only behaviorally relevant information is extracted from the input. The navigation scheme works succesfully in real-time with image resolutions as poor as 322 pixels.

Reliable reaction [mobile robot navigation]

Abstract: In this paper we are concerned with the reliability, i.e., probability of success, of reactive system components (behaviors) in autonomous systems such as mobile robots. A method is presented for designing reliable behaviors by a suitable integration of a set of less reliable ones, which can be based on different and complementary algorithms, sensors, etc. The presented approach is based on the exploitation of the redundancy introduced by the functionally equivalent behaviors, that are committed to pursuing a common goal. Each behavior votes for a set of possible actions and a voter selects the action that best fulfils the goal of the behavior team. We show and experimentally validate that, with careful design, the behavior team will have an improved reliability compared to any of the behaviors constituting it. Real-world experiments are presented, where a particular team of obstacle avoidance behaviors for mobile robot navigation is studied.

View-time based moving obstacle avoidance using stochastic prediction of obstacle motion

Abstract: This paper proposes a new motion planning method of a mobile robot avoiding moving obstacles. To avoid moving obstacles, the trajectories of the obstacles are predicted using a stochastic model of obstacle motion. The obstacle motion is modeled as a random walk process. The method plans robot motion by the unit of view-time and view-period. View-time is defined as the time instant at which the robot senses the obstacle position and velocity. View-period is defined as the time interval during which the robot performs sensing, predicting and planning for collision-free motion. From the position and velocity at a view-time, we predict the future position of the obstacle. The random walk process model of obstacle motion is used to calculate the probability density that the predicted position is reached during the view-period. From the probability density function of the predicted position, the probability that a position can be swept by the obstacle during the view-period is calculated. Then artificial potential is assigned at every position by considering the probability. The force induced by the artificial potential field repels the robot away from the probable obstacle trajectory. This method is a look ahead scheme, and effective for moving obstacle avoidance. This method is applied to collision-free motion planning for a mobile robot in a dynamic and unknown environment with several moving and stationary obstacles.

A sensor-based obstacle avoidance for a redundant manipulator using a velocity potential function

Abstract: A new approach based on artificial potential function is proposed for the obstacle avoidance of redundant manipulators. Unlike the so-called "global" path planning method, which requires expensive computations for the path search before the manipulator starts to move, this new approach, called the "local" path planning, searches the path in real-time using the local distance information. Previous use of artificial potential functions has exhibited local minima in some complex environments. This paper proposes a potential function that has no local minima even for a cluttered environment. The proposed potential function has been implemented for the collision avoidance of a redundant robot in simulation. A simulation is demonstrated on an algorithm that prevents collisions with obstacles by calculating the repulsive potential exerted on links, based on the shortest distance to an object.

A motion planning method for an AUV

Abstract: The authors have been developing an underwater vehicle "Umihico" that autonomously plans and executes given missions. This paper discusses the problem of finding an optimal motion plan for an AUV. The proposed motion planning algorithm is divided into two steps, the global path planning and the local motion planning steps. In the global path planning step, we first define a connectivity graph to evaluate the cost of each path class. Next we apply the Dijkstra's algorithm or the all-pairs cost algorithm to the graph to find the optimal (minimum cost) path class. In the local motion planning step, a smooth path segment which connects two configurations in the optimal path class in each region is computed. If there are any obstacles, the local motion planning algorithm will plan and execute an obstacle avoiding action. The effectiveness and robustness of the solution algorithm are verified through simulation.