Learning to track colored objects with log-polar vision

Metta, Giorgio; Γαστεράτος, Αντώνιος; Sandini, G.

doi:10.1016/j.mechatronics.2004.05.003

Cited by 37 publications

(19 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We use an image processing scheme similar to the one proposed in Metta et al (2004). We convert 320×240 images streamed from the cameras to 150×150 log-polar images.…”

Section: Robot Vision Systemmentioning

confidence: 99%

Learning robotic eye–arm–hand coordination from human demonstration: a coupled dynamical systems approach

Lukic

Santos-Victor²,

Billard

2014

Biol Cybern

View full text Add to dashboard Cite

We investigate the role of obstacle avoidance in visually guided reaching and grasping movements. We report on a human study in which subjects performed prehensile motion with obstacle avoidance where the position of the obstacle was systematically varied across trials. These experiments suggest that reaching with obstacle avoidance is organized in a sequential manner, where the obstacle acts as an intermediary target. Furthermore, we demonstrate that the notion of workspace travelled by the hand is embedded explicitly in a forward planning scheme, which is actively involved in detecting obstacles on the way when performing reaching. We find that the gaze proactively coordinates the pattern of eye-arm motion during obstacle avoidance. This study provides also a quantitative assessment of the coupling between the eye-arm-hand motion. We show that the coupling follows regular phase dependencies and is unaltered during obstacle avoidance. These observations provide a basis for the design of a computational model. Our controller extends the coupled dynamical systems framework and provides fast and synchronous control of the eyes, the arm and the hand within a single and compact framework, mimicking similar control system found in humans. We validate our model for visuomotor control of a humanoid robot.

show abstract

“…We use an image processing scheme similar to the one proposed in Metta et al (2004). We convert 320×240 images streamed from the cameras to 150×150 log-polar images.…”

Section: Robot Vision Systemmentioning

confidence: 99%

Learning robotic eye–arm–hand coordination from human demonstration: a coupled dynamical systems approach

Lukic

Santos-Victor²,

Billard

2014

Biol Cybern

View full text Add to dashboard Cite

show abstract

“…As shown in in Fig.5. The position of object center 1 C in last frame is ( 1 x , 1 y ), the position of object center 2 C in the current frame is ( 2 x , 2 y ), the position of object proper center 3 C′ in the next frame is ( 3 x′ , 3 y′ ). The duration between last frame and current frame is 1 t , and the duration between current frame and next frame is 2 t .…”

Section: Figure4 Tracking Object Situation and Miss Object Situationmentioning

confidence: 99%

“…Tracking moving object is a hot subject in the field of machine vision [1][2][3] . This technology has a lot of potentials in industry, agriculture, and national defence.…”

Section: Introductionmentioning

confidence: 99%

Research on the Method of Tracking and Measuring Moving Object Based on Machine Vision

Zhao¹,

Cui²

2010

2010 International Conference on Computational Intelligence and Software Engineering

View full text Add to dashboard Cite

Abstract-Aiming at the problem of tracking moving object under complex background, a new method is proposed, which is based on machine vision. This method takes advantage of image analysis technology to detect the profile of specified moving object and compute its center. According to the center of moving object in the several previous frame image, predict the possible center in the next frame. Take this predicted center as the input to compute the real center in the next frame. Making this process again and again, ultimately, the moving object's position in every frame is obtained, so that the aim of tracking moving object is achieved. Experimental results are shown here to demonstrate the effectiveness of the proposed method.

show abstract

“…A functional module with an integrated camera is designed and the module is compatible with M-TRAN; docking is realized through visual feedback in a special docking configuration [22]. A method to control the movement of the robotic head is designed, in which colour information is used to determine the position of the object in motion [23].…”

Section: Introductionmentioning

confidence: 99%

A Multi-Sensory Autonomous Docking Approach for a Self-Reconfigurable Robot without Mechanical Guidance

Zhu

Jin

Zhang

et al. 2014

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

The most important feature of a Self-Reconfigurable Robot (SRR) is that it is reconfigurable and self-repairing. At the centre of these capabilities is autonomous docking. One difficulty for docking is the alignment between two robots. Current strategies overcome this by integrating a mechanical guiding device within the connecting mechanism. This increases the robustness of docking but compromises the flexibility of reconfiguration. In this paper, we present a new autonomous docking strategy that can overcome the drawbacks of current approaches. The new strategy uses a novel hook-type connecting mechanism and multi-sensory guidance. The hook-type connecting mechanism is strong and rigid for reliable physical connection between the modules. The multisensory docking strategy, which includes visual-sensorguided rough positioning, Hall-sensor-guided fine positioning, and the locking between moving and target modules, guarantees robust docking without sacrificing reconfigurability. The proposed strategy is verified by docking between a worm-shaped robot and one target module, and docking among three moving robots to form a T-shaped configuration. The experimental results showed that the strategy is very effective.

show abstract

Learning to track colored objects with log-polar vision

Cited by 37 publications

References 36 publications

Learning robotic eye–arm–hand coordination from human demonstration: a coupled dynamical systems approach

Learning robotic eye–arm–hand coordination from human demonstration: a coupled dynamical systems approach

Research on the Method of Tracking and Measuring Moving Object Based on Machine Vision

A Multi-Sensory Autonomous Docking Approach for a Self-Reconfigurable Robot without Mechanical Guidance

Contact Info

Product

Resources

About