Dynamic visual servo control of a 4-axis joint tool to track image trajectories during machining complex shapes

A large part of the new generation of computer numerical control systems has adopted an architecture based on robotic systems. This architecture improves the implementation of many manufacturing processes in terms of flexibility, efficiency, accuracy and velocity. This paper presents a 4-axis robot tool based on a joint structure whose primary use is to perform complex machining shapes in some non-contact processes. A new dynamic visual controller is proposed in order to control the 4-axis joint structure, where image information is used in the control loop to guide the robot tool in the machining task. In addition, this controller eliminates the chaotic joint behavior which appears during tracking of the quasi-repetitive trajectories required in machining processes. Moreover, this robot tool can be coupled to a manipulator robot in order to form a multi-robot platform for complex manufacturing tasks. Therefore, the robot tool could perform a machining task using a piece grasped from the workspace by a manipulator robot. This manipulator robot could be guided by using visual information given by the robot tool, thereby obtaining an intelligent multi-robot platform controlled by only one camera.

Visual Feedforward Control in Human Locomotion During Avoidance of Obstacles that Change Size

The purpose of the current study was to understand how visual information about an ongoing change in obstacle size is used during obstacle avoidance for both lead and trail limbs. Participants were required to walk in a dark room and to step over an obstacle edged with a special tape visible in the dark. The obstacle's dimensions were manipulated one step before obstacle clearance by increasing or decreasing its size. Two increasing and two decreasing obstacle conditions were combined with seven control static conditions. Results showed that information about the obstacle's size was acquired and used to modulate trail limb trajectory, but had no effect on lead limb trajectory. The adaptive step was influenced by the time available to acquire and process visual information. In conclusion, visual information about obstacle size acquired during lead limb crossing was used in a feedforward manner to modulate trail limb trajectory.

Mecanismos de Fotopercepción No-Visual y Control Circadiano en la Retina Interna de Vertebrados en Condiciones Fisiológicas y en un Modelo Animal de Ceguera. Mechanisms of Non-visual Perception and Circadian Control in the Vertebrate Inner Retina under Physiological Conditions and in Animal Model of Blindness.

La retina juega un rol esencial en el funcionamiento del sistema circadiano de los vertebrados al ser la encargada de sensar las condiciones de iluminación ambiental que ajustan el reloj interno con el fotoperíodo exterior a través de un circuito no-visual. Este circuito es independiente de la vía de formación de imágenes e involucra a las células ganglionares retinianas (CGRs) que proyectan a varias estructuras no-visuales del cerebro; esta vía es la encargada de regular el reflejo pupilar, la sincronización de los ritmos diarios de actividad, el sueño y la supresión de melatonina pineal. La retina contiene además un reloj autónomo que genera ritmos diarios autosostenidos en distintas funciones bioquímicas y fisiológicas, que le confiere la capacidad de predecir el tiempo y anticiparse en su fisiología a los cambios lumínicos a lo largo del ciclo día-noche. Este laboratorio ha demostrado por 1ra vez que las CGRs de pollo poseen osciladores endógenos que generan variaciones diarias en la biosíntesis de fosfolípidos (Guido et al, J Neurochem. 2001; Garbarino et al., J Neurosci Res. 2004a) y de la hormona melatonina con niveles máximos durante el día (Garbarino et al., J Biol Chem 2004b). Aún más, cultivos primarios de CGRs responden a la luz a través de una cascada bioquímica de fototransducción similar a la de invertebrados y que involucra la activación de la enzima fosfolipasa C (PLC) (Contin et al., FASEB J 2006). Estos cultivos fueron obtenidos a estadios embrionarios muy tempranos en dónde solo las CGRs son postmitóticas y mayoritariamente maduras. A estos estadios, los cultivos expresan marcadores de especificación de células ganglionares (pax6, brn3), la proteina Gq y los fotopigmentos melanopsina y criptocromos con gran homología con marcadores descriptos para fotorreceptores rabdoméricos de invertebrados (Contin et al, 2006). Recientemente comenzamos a investigar la percepción de luz en pollos GUCY1*, un modelo de ceguera, en animales que carecen de células fotorreceptoras-conos y bastones-funcionales. Resultados preliminares indicarían que la retina interna, y potencialmente las CGRs de estos animales conservarían la capacidad de responder a la luz regulando el reflejo pupilar y sincronizando los ritmos diarios de alimentación. La convergencia de osciladores y fotopigmentos en la población de CGRs podría contribuir al control temporal de la fisiología del organismo y regulación de funciones no-visuales. Son objetivos de este proyecto: a) Investigar el rol de las CGRs en el sistema circadiano estudiando: i- su habilidad para sintetizar melatonina y, su regulación por luz y dopamina; ii- su capacidad fotorreceptora intrínseca, investigando la presencia de fotopigmentos y componentes de la cascada de fototransducción fundamentalmente la vía de los fosfoinosítidos y la activación de PLC, mediante ensayos moleculares, bioquímicos y farmacológicos; b) Extender estos estudios a cultivos primarios de CGRs inmunopurificadas midiendo la respuesta a la luz sobre la síntesis de melatonina, y los niveles de los mensajeros 2rios Ca2+ y AMP cíclico, la inducción de genes tempranos y la regulación de la actividad NAT, enzima clave en la síntesis de melatonina; y c) Investigar la percepción de luz en pollos GUCY1*(ciegos), sobre distintas funciones no-visuales tales como el reflejo pupilar, la sincronización de los ritmos diarios de alimentación, la síntesis de melatonina y la expresión génica en animales expuestos a estimulación lumínica de distintas intensidades y longitudes de onda. Estos estudios permitirán construir el espectro de acción de la respuesta a la luz en los pollos ciegos a fin de identificar el/los fotopigmentos intervinientes en este fenómeno. Este proyecto profundizará el conocimiento sobre la capacidad fotorreceptora-no visual de la retina interna y particularmente de las CGRs, de la naturaleza de la cascada bioquímica que opera en las mismas y de los mecanismos de regeneración del cromóforo utilizado.

Towards Direct Policy Search Reinforcement Learning for Robot Control

This paper proposes a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot. Although the dominant approach, when using RL, has been to apply value function based algorithms, the system here detailed is characterized by the use of direct policy search methods. Rather than approximating a value function, these methodologies approximate a policy using an independent function approximator with its own parameters, trying to maximize the future expected reward. The policy based algorithm presented in this paper is used for learning the internal state/action mapping of a behavior. In this preliminary work, we demonstrate its feasibility with simulated experiments using the underwater robot GARBI in a target reaching task

Intelligent task-level grasp mapping for robot control

In the future, robots will enter our everyday lives to help us with various tasks.For a complete integration and cooperation with humans, these robots needto be able to acquire new skills. Sensor capabilities for navigation in real humanenvironments and intelligent interaction with humans are some of the keychallenges.Learning by demonstration systems focus on the problem of human robotinteraction, and let the human teach the robot by demonstrating the task usinghis own hands. In this thesis, we present a solution to a subproblem within thelearning by demonstration field, namely human-robot grasp mapping. Robotgrasping of objects in a home or office environment is challenging problem.Programming by demonstration systems, can give important skills for aidingthe robot in the grasping task.The thesis presents two techniques for human-robot grasp mapping, directrobot imitation from human demonstrator and intelligent grasp imitation. Inintelligent grasp mapping, the robot takes the size and shape of the object intoconsideration, while for direct mapping, only the pose of the human hand isavailable.These are evaluated in a simulated environment on several robot platforms.The results show that knowing the object shape and size for a grasping taskimproves the robot precision and performance

JXTA-Overlay P2P Platform and Its Application for Robot Control

The design of an efficient collaborative multirobot framework that ensures the autonomy and the individualrequirements of the involved robots is a very challenging task. This requires designing an efficient platform for inter-robot communication. P2P is a good approach to achieve this goal. P2P aims at making the communication ubiquitous thereby crossing the communication boundary and has many attractive features to use it as a platform for collaborative multi-robot environments. In this work, we present the JXTA Overlay P2P system and its application for robot control. Since JXTAOverlay is able to overcome Firewalls, Routers and NATs, it is possible to control end-devices in a WAN without changing the network security policy. We used JXTA-Overlay for the control of robot motors. We evaluated the proposed system by many experiments and have shown that the proposed system has a good performance and can be used successfully for the control of robot.

Towards Direct Policy Search Reinforcement Learning for Robot Control

This paper proposes a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot. Although the dominant approach, when using RL, has been to apply value function based algorithms, the system here detailed is characterized by the use of direct policy search methods. Rather than approximating a value function, these methodologies approximate a policy using an independent function approximator with its own parameters, trying to maximize the future expected reward. The policy based algorithm presented in this paper is used for learning the internal state/action mapping of a behavior. In this preliminary work, we demonstrate its feasibility with simulated experiments using the underwater robot GARBI in a target reaching task

Transputer based real time robot control

A parallel processor architecture based on a communicating sequential processor chip, the transputer, is described. The architecture is easily linearly extensible to enable separate functions to be included in the controller. To demonstrate the power of the resulting controller some experimental results are presented comparing PID and full inverse dynamics on the first three joints of a Puma 560 robot. Also examined are some of the sample rate issues raised by the asynchronous updating of inertial parameters, and the need for full inverse dynamics at every sample interval is questioned.

Intermittently-visual tracking experiments reveal the roles of error-correction and predictive mechanisms in the human visual-motor control system

Prediction mechanism is necessary for human visual motion to compensate a delay of sensory-motor system. In a previous study, “proactive control” was discussed as one example of predictive function of human beings, in which motion of hands preceded the virtual moving target in visual tracking experiments. To study the roles of the positional-error correction mechanism and the prediction mechanism, we carried out an intermittently-visual tracking experiment where a circular orbit is segmented into the target-visible regions and the target-invisible regions. Main results found in this research were following. A rhythmic component appeared in the tracer velocity when the target velocity was relatively high. The period of the rhythm in the brain obtained from environmental stimuli is shortened more than 10%. The shortening of the period of rhythm in the brain accelerates the hand motion as soon as the visual information is cut-off, and causes the precedence of hand motion to the target motion. Although the precedence of the hand in the blind region is reset by the environmental information when the target enters the visible region, the hand motion precedes the target in average when the predictive mechanism dominates the error-corrective mechanism.

Bio-inspired autonomous visual vertical control of a quadrotor UAV

Near ground maneuvers, such as hover, approach and landing, are key elements of autonomy in unmanned aerial vehicles. Such maneuvers have been tackled conventionally by measuring or estimating the velocity and the height above the ground often using ultrasonic or laser range finders. Near ground maneuvers are naturally mastered by flying birds and insects as objects below may be of interest for food or shelter. These animals perform such maneuvers efficiently using only the available vision and vestibular sensory information. In this paper, the time-to-contact (Tau) theory, which conceptualizes the visual strategy with which many species are believed to approach objects, is presented as a solution for Unmanned Aerial Vehicles (UAV) relative ground distance control. The paper shows how such an approach can be visually guided without knowledge of height and velocity relative to the ground. A control scheme that implements the Tau strategy is developed employing only visual information from a monocular camera and an inertial measurement unit. To achieve reliable visual information at a high rate, a novel filtering system is proposed to complement the control system. The proposed system is implemented on-board an experimental quadrotor UAV and shown not only to successfully land and approach ground, but also to enable the user to choose the dynamic characteristics of the approach. The methods presented in this paper are applicable to both aerial and space autonomous vehicles.

Bioinspired autonomous visual vertical control of a quadrotor unmanned aerial vehicle

Near-ground maneuvers, such as hover, approach, and landing, are key elements of autonomy in unmanned aerial vehicles. Such maneuvers have been tackled conventionally by measuring or estimating the velocity and the height above the ground, often using ultrasonic or laser range finders. Near-ground maneuvers are naturally mastered by flying birds and insects because objects below may be of interest for food or shelter. These animals perform such maneuvers efficiently using only the available vision and vestibular sensory information. In this paper, the time-tocontact (tau) theory, which conceptualizes the visual strategy with which many species are believed to approach objects, is presented as a solution for relative ground distance control for unmanned aerial vehicles. The paper shows how such an approach can be visually guided without knowledge of height and velocity relative to the ground. A control scheme that implements the tau strategy is developed employing only visual information from a monocular camera and an inertial measurement unit. To achieve reliable visual information at a high rate, a novel filtering system is proposed to complement the control system. The proposed system is implemented onboard an experimental quadrotor unmannedaerial vehicle and is shown to not only successfully land and approach ground, but also to enable the user to choose the dynamic characteristics of the approach. The methods presented in this paper are applicable to both aerial and space autonomous vehicles.

Situation assessment through multi-modal sensing of dynamic environments to support cognitive robot control

Awareness of emerging situations in a dynamic operational environment of a robotic assistive device is an essential capability of such a cognitive system, based on its effective and efficient assessment of the prevailing situation. This allows the system to interact with the environment in a sensible (semi)autonomous / pro-active manner without the need for frequent interventions from a supervisor. In this paper, we report a novel generic Situation Assessment Architecture for robotic systems directly assisting humans as developed in the CORBYS project. This paper presents the overall architecture for situation assessment and its application in proof-of-concept Demonstrators as developed and validated within the CORBYS project. These include a robotic human follower and a mobile gait rehabilitation robotic system. We present an overview of the structure and functionality of the Situation Assessment Architecture for robotic systems with results and observations as collected from initial validation on the two CORBYS Demonstrators.