Task-Level Robot Learning

We are investigating how to program robots so that they learn from experience. Our goal is to develop principled methods of learning that can improve a robot's performance of a wide range of dynamic tasks. We have developed task-level learning that successfully improves a robot's performance of two complex tasks, ball-throwing and juggling. With task- level learning, a robot practices a task, monitors its own performance, and uses that experience to adjust its task-level commands. This learning method serves to complement other approaches, such as model calibration, for improving robot performance.

Design and Implementation of a Flexible Robot

This robot has low natural frequencies of vibration. Insights into the problems of designing joint and link flexibility are discussed. The robot has three flexible rotary actuators and two flexible, interchangeable links, and is controlled by three independent processors on a VMEbus. Results from experiments on the control of residual vibration for different types of robot motion are presented. Impulse prefiltering and slowly accelerating moves are compared and shown to be effective at reducing residual vibration.

On Probabilistic Strategies for Robot Tasks

Robots must act purposefully and successfully in an uncertain world. Sensory information is inaccurate or noisy, actions may have a range of effects, and the robot's environment is only partially and imprecisely modeled. This thesis introduces active randomization by a robot, both in selecting actions to execute and in focusing on sensory information to interpret, as a basic tool for overcoming uncertainty. An example of randomization is given by the strategy of shaking a bin containing a part in order to orient the part in a desired stable state with some high probability. Another example consists of first using reliable sensory information to bring two parts close together, then relying on short random motions to actually mate the two parts, once the part motions lie below the available sensing resolution. Further examples include tapping parts that are tightly wedged, twirling gears before trying to mesh them, and vibrating parts to facilitate a mating operation.

The PHD: A Planar, Harmonic Drive Robot for Joint Torque Control

This thesis details the development of a model of a seven degree of freedom manipulator for position control. Then, it goes on to discuss the design and construction of a the PHD, a robot built to serve two purposes: first, to perform research on joint torque control schemes, and second, to determine the important dynamic characteristics of the Harmonic Drive. The PHD, is a planar, three degree of freedom arm with torque sensors integral to each joint. Preliminary testing has shown that a simple linear spring model of the Harmonic Drive's flexibility is suitable in many situations.

Modeling, Estimation, and Control of Robot-Soil Interactions

This thesis presents the development of hardware, theory, and experimental methods to enable a robotic manipulator arm to interact with soils and estimate soil properties from interaction forces. Unlike the majority of robotic systems interacting with soil, our objective is parameter estimation, not excavation. To this end, we design our manipulator with a flat plate for easy modeling of interactions. By using a flat plate, we take advantage of the wealth of research on the similar problem of earth pressure on retaining walls. There are a number of existing earth pressure models. These models typically provide estimates of force which are in uncertain relation to the true force. A recent technique, known as numerical limit analysis, provides upper and lower bounds on the true force. Predictions from the numerical limit analysis technique are shown to be in good agreement with other accepted models. Experimental methods for plate insertion, soil-tool interface friction estimation, and control of applied forces on the soil are presented. In addition, a novel graphical technique for inverting the soil models is developed, which is an improvement over standard nonlinear optimization. This graphical technique utilizes the uncertainties associated with each set of force measurements to obtain all possible parameters which could have produced the measured forces. The system is tested on three cohesionless soils, two in a loose state and one in a loose and dense state. The results are compared with friction angles obtained from direct shear tests. The results highlight a number of key points. Common assumptions are made in soil modeling. Most notably, the Mohr-Coulomb failure law and perfectly plastic behavior. In the direct shear tests, a marked dependence of friction angle on the normal stress at low stresses is found. This has ramifications for any study of friction done at low stresses. In addition, gradual failures are often observed for vertical tools and tools inclined away from the direction of motion. After accounting for the change in friction angle at low stresses, the results show good agreement with the direct shear values.

Semi-online neural-Q_leaming for real-time robot learning

Reinforcement learning (RL) is a very suitable technique for robot learning, as it can learn in unknown environments and in real-time computation. The main difficulties in adapting classic RL algorithms to robotic systems are the generalization problem and the correct observation of the Markovian state. This paper attempts to solve the generalization problem by proposing the semi-online neural-Q_learning algorithm (SONQL). The algorithm uses the classic Q_learning technique with two modifications. First, a neural network (NN) approximates the Q_function allowing the use of continuous states and actions. Second, a database of the most representative learning samples accelerates and stabilizes the convergence. The term semi-online is referred to the fact that the algorithm uses the current but also past learning samples. However, the algorithm is able to learn in real-time while the robot is interacting with the environment. The paper shows simulated results with the "mountain-car" benchmark and, also, real results with an underwater robot in a target following behavior

VLSI architecture for an Underwater Robot Vision System

It is well known that image processing requires a huge amount of computation, mainly at low level processing where the algorithms are dealing with a great number of data-pixel. One of the solutions to estimate motions involves detection of the correspondences between two images. For normalised correlation criteria, previous experiments shown that the result is not altered in presence of nonuniform illumination. Usually, hardware for motion estimation has been limited to simple correlation criteria. The main goal of this paper is to propose a VLSI architecture for motion estimation using a matching criteria more complex than Sum of Absolute Differences (SAD) criteria. Today hardware devices provide many facilities for the integration of more and more complex designs as well as the possibility to easily communicate with general purpose processors

PRIM an Open Mobile Robot Platform. Motivation, Present and Future Trends

This paper is focused on the robot mobile platform PRIM (platform robot information multimedia). This robot has been made in order to cover two main needs of our group, on one hand the need for a full open mobile robotic platform that is very useful in fulfilling the teaching and research activity of our school community, and on the other hand with the idea of introducing an ethical product which would be useful as mobile multimedia information point as a service tool. This paper introduces exactly how the system is made up and explains just what the philosophy is behind this work. The navigation strategies and sensor fusion, where machine vision system is the most important one, are oriented towards goal achievement and are the key to the behaviour of the robot

Local WMR navigation with monocular data

This article presents recent WMR (wheeled mobile robot) navigation experiences using local perception knowledge provided by monocular and odometer systems. A local narrow perception horizon is used to plan safety trajectories towards the objective. Therefore, monocular data are proposed as a way to obtain real time local information by building two dimensional occupancy grids through a time integration of the frames. The path planning is accomplished by using attraction potential fields, while the trajectory tracking is performed by using model predictive control techniques. The results are faced to indoor situations by using the lab available platform consisting in a differential driven mobile robot

Local model predictive control experiences with differential driven wheeled mobile robots

This work extends a previously developed research concerning about the use of local model predictive control in differential driven mobile robots. Hence, experimental results are presented as a way to improve the methodology by considering aspects as trajectory accuracy and time performance. In this sense, the cost function and the prediction horizon are important aspects to be considered. The aim of the present work is to test the control method by measuring trajectory tracking accuracy and time performance. Moreover, strategies for the integration with perception system and path planning are briefly introduced. In this sense, monocular image data can be used to plan safety trajectories by using goal attraction potential fields

A Multi-agent Architecture with Distributed Coordination for an Autonomous Robot

Aquest treball proposa una nova arquitectura de control amb coordinació distribuïda per a un robot mòbil (ARMADiCo). La metodologia de coordinació distribuïda consisteix en dos passos: el primer determina quin és l'agent que guanya el recurs basat en el càlcul privat de la utilitat i el segon, com es fa el canvi del recurs per evitar comportaments abruptes del robot. Aquesta arquitectura ha estat concebuda per facilitar la introducció de nous components hardware i software, definint un patró de disseny d'agents que captura les característiques comunes dels agents. Aquest patró ha portat al desenvolupament d'una arquitectura modular dins l'agent que permet la separació dels diferents mètodes utilitzats per aconseguir els objectius, la col·laboració, la competició i la coordinació de recursos. ARMADiCo s'ha provat en un robot Pioneer 2DX de MobileRobots Inc.. S'han fet diversos experiments i els resultats han demostrat que s'han aconseguit les característiques proposades per l'arquitectura.

Control of a compliant two-axis robotic arm

In order to ease control, the links between actuators and robotic limbs are generally made to be as stiff as possible. This is in contrast to natural limbs, where compliance is present. Springs have been added to the drive train between the actuator and load to imitate this natural compliance. The majority of these springs have been in series between the actuator and load. However, a more biologically inspired approach is taken, here springs have been used in parallel to oppose each other. The paper will describe the application of parallel extension springs in a robot arm in order to give it compliance. Advantages and disadvantages of this application are discussed along with various control strategies.

Analysis of the Fugl-Meyer outcome measures assessing the effectiveness of robot-mediated stroke therapy

Robot-mediated therapies offer a new approach to neurorehabilitation. This paper analyses the Fugl-Meyer data from the Gentle/S project and finds that the two intervention phases (sling suspension and robot mediated therapy) have approximately equal value to the further recovery of chronic stroke subjects (on average 27 months post stroke). Both sling suspension and robot mediated interventions show a recovery over baseline and further work is needed to establish the common factors in treatment, and to establish intervention protocols for each that will give individual subjects a maximum level of recovery.

Closely coupled behaviour in networked robot systems

Multiple cooperating robot systems may be required to take up a closely coupled configuration in order to perform a task. An example is extended baseline stereo (EBS), requiring that two robots must establish and maintain for a certain period of time a constrained kinematic relationship to each other. In this paper we report on the development of a networked robotics framework for modular, distributed robot systems that supports the creation of such configurations. The framework incorporates a query mechanism to locate modules distributed across the two robot systems. The work presented in this paper introduces special mechanisms to model the kinematic constraint and its instantiation. The EBS configuration is used as a case study and experimental implementation to demonstrate the approach.

Control of a compliant two-axis robotic arm

It is common to make the links between actuators and robotic limbs as stiff as possible, in complete contrast to natural systems, where compliance is present. In the past, to create some compliance in a drive, springs have been added to the link between the actuator and load. Many of these springs have been in series with the drive, but recently a more 'biological' approach has been taken where two springs have been used in parallel to counteract each other. This paper describes the application of parallel extension springs in a robot arm in order to give it compliance. Advantages and disadvantages of this application are discussed, along with various control strategies.