机器人足球——智能机器人的新领域



机器人足球——智能机器人的新领域

高大志，张春晖，徐心和

摘要：本文主要介绍近年来发展的机器人足球比赛和足球机器人系统。足球机器人系统由机器人、视觉系统、主机系统和通讯系统组成。多机器人组队比赛便构成一个复杂大系统，它涉及机器人、自动控制、通信、传感、图像处理以及人工智能等领域，而机器人足球便成为研究多机器人合作以及多智能体系统的一个很好的实验载体。

关键词：机器人；多智能体系统；足球机器人；移动机器人

来源出版物：机器人, 1998, 20(4): 309-314

被引频次：1869

A robust layered control system for a mobile robot

Brooks, RA

Abstract:A new architecture for controlling mobile robots is described. Layers of control system are built to let the robot operate at increasing levels of competence. Layers are made up of asynchronous modules that communicate over low-bandwidth channels. Each module is an instance of a fairly simple computational machine. Higher-level layers can subsume the roles of lower levels by suppressing their outputs. However, lower levels continue to function as higher levels are added. The result is a robust and flexible robot control system. The system has been used to control a mobile robot wandering around unconstrained laboratory areas and computer machine rooms. Eventually it is intended to control a robot that wanders the office areas of our laboratory, building maps of its surroundings using an onboard arm to perform simple tasks. This paper presents a unique real-time obstacle avoidance approach for manipulators and mobile robots based on the artificial potential field concept. Collision avoidance, traditionally considered a high level planning problem, can be effectively distributed between different levels of control, al lowing real-time robot operations in a complex environment. This method has been extended to moving obstacles by using a time-varying artificial potential field. We have applied this obstacle avoidance scheme to robot arm mechanisms and have used a new approach to the general problem of real-time manipulator control. We reformulated the manipulator con trol problem as direct control of manipulator motion in operational space—the space in which the task is originally described—rather than as control of the task's corresponding joint space motion obtained only after geometric and kinematic transformation. Outside the obstacles' regions of influence, we caused the end effectors to move in a straight line with an upper speed limit. The artificial potential field approach has been extended to collision avoidance for all manipulator links. In addition, a joint space artificial potential field is used to satisfy the manipulator internal joint constraints. This method has been implemented in the COSMOS system for a PUMA 560 robot. Real-time collision avoidance demonstrations on moving obstacles have been performed by using visual sensing. This article proposes a betterment process for the operation of a mechanical robot in a sense that it betters the next operation of a robot by using the previous operation's data. The process has an iterative learning structure such that the (k + 1)th input to joint actuators consists of the kth input plus an error increment composed of the derivative difference between the kth motion trajectory and the given desired motion trajectory. The convergence of the process to the desired motion trajectory is assured under some reasonable conditions. Numerical results by computer simulation are presented to show the effectiveness of the proposed learning scheme. A framework for the analysis and control of manipulator systems with respect to the dynamic behavior of their end-effectors is developed. First, issues related to the description of end-effector tasks that involve constrained motion and active force control are discussed. The fundamentals of the operational space formulation are then presented, and the unified approach for motion and force control is developed. The extension of this formulation to redundant manipulator systems is also presented, constructing the end-effector equations of motion and describing their behavior with respect to joint forces. These results are used in the development of a new and systematic approach for dealing with the problems arising at kinematic singularities. At a singular configuration, the manipulator is treated as a mechanism that is redundant with respect to the motion of the end-effector in the subspace of operational space orthogonal to the singular direction. A new real-time obstacle avoidance method for mobile robots has been developed and implemented. This method, named the vector field histogram (VFH), permits the detection of unknown obstacles and avoids collisions while simultaneously steering the mobile robot toward the target. The VFH method uses a two-dimensional Cartesian histogram grid as a world model. This world model is updated continuously with range data sampled by on-board range sensors. The VFH method subsequently employs a two-stage data-reduction process in order to compute the desired control commands for the vehicle. In the first stage the histogram grid is reduced to a one-dimensional polar histogram that is constructed around the robot's momentary location. Each sector in the polar histogram contains a value representing the polar obstacle density in that direction. In the second stage, the algorithm selects the most suitable sector from among all polar histogram sectors with a low polar obstacle density, and the steering of the robot is aligned with that direction. Experimental results from a mobile robot traversing densely cluttered obstacle courses in smooth and continuous motion and at an average speed of 0.6-0.7 m/s demonstrate the power of the VFH method. We present a new methodology for exact robot motion planning and control that unifies the purely kinematic path planning problem with the lower level feedback controller design. Complete information about the free space and goal is encoded in the form of a special artificial potential function-a navigation function-that connects the kinematic planning problem with the dynamic execution problem in a provably correct fashion. The navigation function automatically gives rise to a bounded-torque feedback controller for the robot's actuators that guarantees collision-free motion and convergence to the destination from almost all intial free configurations. Since navigation functions exist for any robot and obstacle course, our methodology is completely general in principle. However, this paper is mainly concerned with certain constructive techniques for a particular class of motion planning problems. Specifically, we present a formula for navigation functions that guide a point-mass robot in a generalized sphere world. The simplest member of this family is a space obtained by puncturing a disc by an arbitrary number of smaller disjoint discs representing obstacles. The other spaces are obtained from this model by a suitable coordinate transformation that we show how to build. Our constructions exploit these coordinate transformations to adapt a navigation function on the model space to its more geometrically complicated (but topologically equivalent) instances. The formula that we present admits sphere-worlds of arbitrary dimension and is directly applicable to configuration spaces whose forbidden regions can be modeled by such generalized discs. We have implemented these navigation functions on planar scenarios, and simulation results are provided throughout the paper. Passive-dynamic walkers are simple mechanical devices, composed of solid parts connected by joints, that walk stably down a slope. They have no motors or controllers, yet can have remarkably humanlike motions. This suggests that these machines are useful models of human locomotion; however, they cannot walk on level. ground. Here we present three robots based on passive-dynamics, with small active power sources substituted for gravity, which can walk on level ground. These robots use less control and less energy than other powered robots, yet walk more naturally, further suggesting the importance of passive-dynamics in human locomotion. A methodology of feedback control is developed to achieve accurate tracking in a. class of non-linear, time-varying systems in the presence of disturbances and parameter variations. The methodology uses in its idealized form piecewise continuous feedback control, resulting in the state trajectory sliding along a time-varying sliding surface in the state space. This idealized control law achieves perfect tracking; however, non-idealities in its implementation result in the generation of an undesirable high-frequency component in the state trajectory. To rectify this, it is shown how continuous control laws may be used to approximate the discontinuous control law to obtain robust tracking to within a prescribed accuracy without generating undesirable high-frequency signal. The method is applied to the control of a two-link manipulator handling variable loads in a flexible manufacturing system environment. A new adaptive robot control algorithm is derived, which consists of a PD feedback part and a full dynamics feed forward compensation part, with the unknown manipulator and payload parameters being estimated online. The algorithm is computationally simple, because of an effective exploitation of the structure of manipulator dynamics. In particular, it requires neither feedback of jointbook=14,ebook=18accelerations nor inversion of the estimated inertia matrix. The algorithm can also be applied directly in Cartesian space. Mobile robot localization is the problem of determining a robot’s pose from sensor data. This article presents a family of probabilistic localization algorithms known as: Monte Carlo Localization (MCL). MCL algorithms represent a robot's belief by a set of weighted hypotheses (samples), which approximate the posterior under a common Bayesian formulation of the localization problem. Building on the basic MCL algorithm, this article develops a more robust algorithm called Mixture-MCL, which integrates two complimentary ways of generating samples in the estimation. To apply this algorithm to mobile robots equipped with range finders, a kernel density tree is learned that permits fast sampling. Systematic empirical results illustrate the robustness and computational efficiency of the approach. It has been known for some time (Gevarter 1970) that if a flexible structure is controlled by locating every sensor exactly at the actuator it will control, then stable operation is easy to achieve. Nearly all commercial robots are controlled in this way, for this reason. So are most flexible spacecraft. Conversely, when one attempts to control a flexible structure by applying control torques at one end that are based on a sensor at the other end, the problem of achieving stability is severe. Solving it is an essential step for better control in space: the space-shuttle arm is a cogent example. The next generation of industrial robots will also need such control capability, for they will need to be much lighter in weight (to achieve quick response with modest energy), and they will need to achieve greater precision by employing end-point sensing. A set of experiments has been constructed to demonstrate control strategies for a single-link, very flexible manipulator, where the position of one end is to be sensed and precisely positioned by torturing at the other end. The objective of this first set of experiments is to uncover and solve problems related to the control of very flexible manipulators where sensors are not collocated with the actuator. The experimental arrangement described here is also a test bed for new designs for flexible-structure controllers, designs that use insensitive, reduced-order control and adaptive control methods, for example. This paper describes the experimental arrangement, model identification, control design, and first experimental results. Some interesting results are the following. First, good stability can be achieved for such noncollocated systems, and reponse can be achieved that is effectively three times faster than the first natural cantilever period of the system: but a good model of the system dynamics and rather sophisticated control algorithms are essential to doing so. Even then, the system will always be conditionally stable. In addition to the tip sensor, a collocated rate sensor and nearly collocated strain gauges have been found to be very useful for achieving good closed-loop performance, that is, high gain and high band width. Second, there is an ultimate physical limit to achievable response time, namely, the time required for a wave to travel the length of the member. Well-designed controllers can approach this limit. Third, the use of end-point sensing makes less critical the elaborate dynamic conditioning of position-command signals— "model-following" differentiators, feed-forward, and the like—such as are typically needed in presentgeneration robots that use "dead reckoning" in lieu of end-point sensing. With end-point sensing, feedback alone (suitably conditioned) is sufficient to whip the tip to the commanded position and hold it there precisely. Even more important, a shift in, for example, work piece with respect to robot base, no longer produces an error. To determine whether simultaneously recorded motor cortex neurons can be used for real-time device control, rats were trained to position a robot arm to obtain water by pressing a lever. Mathematical transformations, including neural networks, converted multineuron signals into 'neuronal population functions' that accurately predicted lever trajectory. Next, these functions were electronically converted into real-time signals for robot arm control. After switching to this 'neurorobotic' mode, 4 of 6 animals (those with >25 task-related neurons) routinely used these brain-derived signals to position the robot arm and obtain water. With continued training in neurorobotic mode, the animals' lever movement diminished or stopped. These results suggest a possible means for movement restoration in paralysis patients. Chain form systems have recently been introduced to model the kinematics of a class of nonholonomic mechanical systems. The first part of the study is centered on control design and analysis for nonlinear systems which can be converted to the chain form. Solutions to various control problems (open-loop steering, partial or complete state feedback stabilization) are either recalled, generalized, or developed. In particular, globally stabilizing time-varying feedbacks are derived, and a discussion of their convergence properties is provided. Application to the control of nonholonomic wheeled mobile robots is described in the second part of the study by considering the case of a car-pulling trailers. In this paper we give a tutorial account of several of the most recent adaptive control results for rigid robot manipulators. Our intent is to lend some perspective to the growing list of adaptive control results for manipulators by providing a unified framework for comparison of those adaptive control algorithms which have been shown to be globally convergent. In most cases we are able to simplify the derivations and proofs of these results as well. In this paper, the authors describe the design and control cf, RHex, a power autonomous, untethered, compliant-legged hexapod robot. RHex has only six actuators-one motor located at each hip-achieving mechanical simplicity that promotes reliable and robust operation in real-world tasks. Empirically stable and highly maneuverable locomotion arises from a very simple clock-driven, open-loop tripod gait. The legs rotate full circle, thereby preventing the common problem of toe stubbing in the protraction (swing) phase. An extensive suite of experimental results documents the robot's significant "intrinsic mobility"-the traversal of rugged, broken, and obstacle-ridden ground without any terrain sensing or actively controlled adaptation. RHex achieves fast and robust forward locomotion traveling at speeds up to one body length per second and traversing height variations well exceeding its body clearance. This paper addresses the control of a team of nonholonomic mobile robots navigating in a terrain with obstacles while maintaining a desired formation and changing formations when required, using graph theory. We model the team as a triple, (g, r, H), consisting of a group element g that describes the gross position of the lead robot, a set of shape variables r that describe the relative positions of robots, and a control graph H that describes the behaviors of the robots in the formation. Our framework enables the representation and enumeration of possible control graphs and the coordination of transitions between any two formations. A multilayer neural-net (NN) controller for a general serial-link rigid robot arm is developed. The structure of the NN controller is derived using a filtered error/passivity approach. No off-line learning phase is needed for the proposed NN controller and the weights are easily initialized. The nonlinear nature of the NN, plus NN functional reconstruction inaccuracies and robot disturbances, mean that thebook=16,ebook=20standard delta rule using backpropagation tuning does not suffice for closed-loop dynamic control. Novel on-line weight tuning algorithms, including correction terms to the delta rule plus an added robustifying signal, guarantee bounded tracking errors as well as bounded NN weights. Specific bounds are determined, and the tracking error bound can be made arbitrarily small by increasing a certain feedback gain. The correction terms involve a second-order forward-propagated wave in the backprop network. New NN properties including the notions of a passive NN, a dissipative NN, and a robust NN are introduced. A real-time obstacle avoidance approach for mobile robots has been developed and implemented. It permits the detection of unknown obstacles simultaneously with the steering of the mobile robot to avoid collisions and advance toward the target. The novelty of this approach, entitled the virtual force field method, lies in the integration of two known concepts: certainty grids for obstacle representation and potential fields for navigation. This combination is especially suitable for the accommodation of inaccurate sensor data as well as for sensor fusion and makes possible continuous motion of the robot with stopping in front of obstacles. This navigation algorithm also takes into account the dynamic behavior of a fast mobile robot and solves the local minimum trap problem. Experimental results from a mobile robot running at a maximum speed of 0.78 m/s demonstrate the power of the algorithm. An approach to robot perception and world modeling that uses a probabilistic tesselated representation of spatial information called the occupancy grid is reviewed. The occupancy grid is a multidimensional random field that maintains stochastic estimates of the occupancy state of the cells in a spatial lattice. To construct a sensor-derived map of the robot's world, the cell state estimates are obtained by interpreting the incoming range readings using probabilistic sensor models. Bayesian estimation procedures allow the incremental updating of the occupancy grid, using readings taken from several sensors over multiple points of view. The use of occupancy grids from mapping and for navigation is examined. Operations on occupancy grids and extensions of the occupancy grid framework are briefly considered. Objective: To compare the effects of robot-assisted movement training with conventional techniques for the rehabilitation of upper-limb motor function after stroke. Design: Randomized controlled trial, 6-month follow-up. Setting: A Department of Veterans Affairs rehabilitation research and development center. Participants: Consecutive sample of 27 subjects with chronic hemiparesis (>6 mo after cerebrovascular accident) randomly allocated to group. Interventions: All subjects received twenty-four I-hour sessions over 2 months. Subjects in the robot group practiced shoulder and elbow movements while assisted by a robot manipulator. Subjects in the control group received neurodevelopmental therapy (targeting proximal upper limb function) and 5 minutes of exposure to the robot in each session. Main Outcome Measures: Fugl-Meyer assessment of motor impairment, FIM(TM) instrument, and biomechanic measures of strength and reaching kinematics. Clinical evaluations were performed by a therapist blinded to group assignments. Results: Compared with the control group, the robot group had larger improvements in the proximal movement portion of the Fugl-Meyer test after 1 month of treatment (P < 0.05) and also after 2 months of treatment (P < 0.05). The robot group had larger gains in strength (P < 0.02) and larger increases in reach extent (P < 0.01) after 2 months of treatment. At the 6-month follow-up, the groups no longer differed in terms of the Fugl-Meyer test (P > 0.30); however, the robot group had larger improvements in the FIM (P < 0.04). Conclusions: Compared with conventional treatment, robot-assisted movements had advantages in terms of clinical and biomechanical measures. Further research into the use of robotic manipulation for motor rehabilitation is justified.

来源出版物： IEEE Journal of Robotics and Automation, 1986, 2(1): 14-23

被引频次：1780

Real-time obstacle avoidance for manipulators and mobile robots

Khatib, O

来源出版物： International Journal of Robotics Research, 1986, 5(1): 90-98

被引频次：1196

Bettering operation of robots by learning

Arimoto, S; Kawamura, S; Miyazaki, F

来源出版物：Journal of Robotic Systems, 1984, 1(2): 123-140

被引频次：769

A unified approach for motion and force control of robot manipulator: The operational space formulation

Khatib, O

来源出版物： IEEE Journal of Robotics and Automation, 1987, 3(1): 43-53

被引频次：697

The vector field histogram - fast obstacle avoidance for mobile robots

Borenstein, J; Koren, Y

Keywords: navigation; sonar

来源出版物： IEEE Journal of Robotics and Automation, 1991, 7(3): 278-288

被引频次：684

Exact robot navigation using artificial potential functions

Rimon, E; Koditschek, DE

Keywords:time obstacle avoidance; mobile robots; manipulators; motion walking; reinforcement; locomotion mobile robots; localization; position estimation; particle filters; kernel density trees somatic sensory transmission; movement; rat; patterns; restoration; modulation; ensembles; thalamus laws adaptive control; robots; parameter estimation; passivity; non-linear control; robustness legged locomotion; hexapod robot; clock driven; mobility; autonomy; biomimesis formation control of mobile robots; graph theory; non-linear control adaptive-control; robust-control; networks; systems mobile robots; localization; position estimation; particle filters; kernel density trees arm; cerebrovascular accident; movement; rehabilitation; robotics; therapy

来源出版物： IEEE Journal of Robotics and Automation, 1992, 8(5): 501-518

被引频次：677

Efficient bipedal robots based on passive-dynamic walkers

Collins, S; Ruina, A; Tedrake, R; et al.

来源出版物： Science, 2005, 307(5712): 1082-1085

被引频次：646

Tracking control of non-linear systems using sliding surfaces, with application to robot manipulators

Slotine, JJ; Sastry, SS

来源出版物: International Journal of Control, 1983, 38(2): 465-492

被引频次：616

On the adaptive-control of robot manipulators

Slotine, JJE; Li, WP

来源出版物：International Journal of Robotics Research, 1987, 6(3): 49-59

被引频次：508

Robust Monte Carlo localization for mobile robots

Thrun, S; Fox, D; Burgard, W; et al.

来源出版物： Artificial Intelligence, 2001, 128(1): 99-141

被引频次：503

Initial experiments on the endpoint control of a flexible one-link robot

Cannon, RH; Schmitz, E

来源出版物： International Journal of Robotics Research, 1984, 3(3): 62-75

被引频次：500

Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex

Chapin, JK; Moxon, KA; Markowitz, RS; et al.

来源出版物： Nature Neuroscience, 1999, 2(7): 664-670

被引频次：495

Control of chained systems application to path-following and time-varying point-stabilization of mobile robots

Samson, C

来源出版物： IEEE Transactions on Automatic Control, 1995, 40(1): 64-77

被引频次：467

Adaptive motion control of rigid robots: A tutorial

Ortega, R; Spong, MW

来源出版物： Automatica, 1989, 25(6): 877-888

被引频次：458

RHex: A simple and highly mobile hexapod robot

Saranli, U; Buehler, M; Koditschek, DE

来源出版物： International Journal of Robotics Research, 2001, 20(7): 616-631

被引频次：448

Modeling and control of formations of nonholonomic mobile robots

Desai, JP; Ostrowski, JP; Kumar, V

来源出版物： IEEE Transactions on Robotics and Automation, 2001, 17(6): 905-908

被引频次：437

Multilayer neural-net robot controller with guaranteed tracking performance

Lewis, F; Yesildirek, A; Liu, K

来源出版物： IEEE Transactions on Neural Networks, 1996, 7(2): 388-399

被引频次：421

Real-time obstacle avoidance for fast mobile robots

Borenstein, J; Koren, Y

来源出版物：IEEE Transactions on Man and Cybernetics, 1989, 19(5): 1179-1187

被引频次：409

Using occupancy grids for mobile robot perception and navigation

Elfes, A

来源出版物： Archives of Physical Medicine and Rehabilitation, 2002, 83(7): 952-959

专家推荐

来源出版物： Computer, 1989, 22(6): 46-57

被引频次：390

Robot-assisted movement training compared with conventional therapy techniques for the rehabilitation of upper-limb motor function after stroke

Lum, PS; Burgar, CG; Shor, PC