Visual control of robotic manipulator based on neural networks

Hashimoto, Hideki; Kubota, Takashi; Sato, M.; Harashima, Fumio

doi:10.1109/41.170967

Cited by 59 publications

(29 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The feedforward signal a a a oi for the image feature acceleration caused by the object's unknown motion is estimated iteratively by (9)- (12). The output of the feedforward controller is actually the sum of the output of the PI controller and the feedforward signal a a a p fi (k) + KF a a aoi (k), where we choose KF = 0:3 and m = 6 empirically in control.…”

Section: A Low-speed Visual Trackingmentioning

confidence: 99%

“…A lot of applications have been accumulated such as robot control [6], chemical process modeling [7], nonlinear system estimation and control [8], image coding [18] and pattern recognition [9], [12], etc. Incremental learning ability from local receptive-field is proved to be extremely useful for approximating unknown functional relationships between input and output data streams [11].…”

Section: Incremental Learning With Balanced Update On Receptive Fieldmentioning

confidence: 99%

“…There also have been ways to estimate the image Jacobian matrix in real-time control of robotic hand-eye coordination system, such as by using the tool of artificial neural networks (ANNs) [11], [12]. The offline training of the adopted ANN extraordinarily reduces computational complexity of online control in comparison with that of online estimation of image Jacobian matrix.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Nonlinear Visual Mapping Model for 3-D Visual Tracking With Uncalibrated Eye-in-Hand Robotic System

Yang

Hanebeck³

et al. 2004

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

Abstract-A new control scheme for uncalibrated robotic visual tracking problem is proposed that compromises the computational expenses of overall system with offline modeling and online control. A nonlinear visual mapping model for the uncalibrated hand-eye coordination is first proposed with an artificial neural network implementation. An online visual tracking controller is then developed together with a real-time motion planner. To improve the system performance, the control scheme is also integrated with a feedforward controller to compensate unknown object motions. Extensive simulations and experiments demonstrate the effectiveness of the proposed control scheme.Index Terms-Hand/eye coordination, image Jacobian matrix, neural networks, nonlinear mapping.

show abstract

Section: A Low-speed Visual Trackingmentioning

confidence: 99%

Section: Incremental Learning With Balanced Update On Receptive Fieldmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Nonlinear Visual Mapping Model for 3-D Visual Tracking With Uncalibrated Eye-in-Hand Robotic System

Yang

Hanebeck³

et al. 2004

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

show abstract

“…Artificial Neural Network (ANN) are nonlinear, global approximation methods. Neural networks have been heavily employed in robotics technology such as robot arm visual control as introduced by Hashimoto, Kubota, Sato, and Harashima [4] , Hashimoto, Kubota, Kuduo, and Harashima [5] , inverse kinematics problem of six degree of freedom robot arm as done by Guez and Ahmed [6] , the research introduced by Patrick and Krose [7] in which they employ a real-time learning neural robot controller for solving the inverse kinematics problem, and the research introduced by Schram, Linden, Krose, and Groen [8] , in which they employ an artificial neural network for tracking and grasping a moving object observed by a six degree of freedom robotics arm system.…”

Section: Multi-fingered Robot Hand and The Dynamics Of Manipulationmentioning

confidence: 99%

Neural Networks for Multi-Finger Robot Hand Control

Mattar¹

2008

eng

View full text Add to dashboard Cite

Abstract. This paper investigates the employment of Artificial Neural Networks (ANN) for a multi-finger robot hand manipulation in which the object motion is defined in task-space with respect to six Cartesian based coordinates. The approach followed here is to let an ANN learn the nonlinear functional relating the entire hand joints positions and displacements to object displacement. This is done by considering the inverse hand Jacobian, in addition to the interaction between hand fingers and the object being grasped and manipulated. The developed network has been trained for several object training patterns and postures within a Cartesian based palm dimension. The paper demonstrates the proposed algorithm for a four fingered robot hand, where inverse hand Jacobian plays an important role in robot hand dynamic control.

show abstract

“…The high computation speed and general modelling capability of neural networks are very attractive properties for nonlinear compensation problems, as indeed robot control problems are. Hashimoto et al utilized two BP networks -one global and one local -to approximate the Jacobian mapping [6,7]; Kuhn et al used an adaptive linear (ADALINE) network to approximate the inverse image Jacobian and a self-organizing map (SOM) network to select the ADALINE matrix [8]. However, the significance of visual servoing is its flexibility, but the nature of offline learning critically weakens this advantage.…”

Section: Introductionmentioning

confidence: 99%

Image Jacobian Matrix Estimation Based on Online Support Vector Regression

Mao

Huang

Wang

2012

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Research into robotics visual servoing is an important area in the field of robotics. It has proven difficult to achieve successful results for machine vision and robotics in unstructured environments without using any a priori camera or kinematic models. In uncalibrated visual servoing, image Jacobian matrix estimation methods can be divided into two groups: the online method and the offline method. The offline method is not appropriate for most natural environments. The online method is robust but rough. Moreover, if the images feature configuration changes, it needs to restart the approximating procedure. A novel approach based on an online support vector regression (OL-SVR) algorithm is proposed which overcomes the drawbacks and combines the virtues just mentioned.

show abstract

Visual control of robotic manipulator based on neural networks

Cited by 59 publications

References 9 publications

Nonlinear Visual Mapping Model for 3-D Visual Tracking With Uncalibrated Eye-in-Hand Robotic System

Nonlinear Visual Mapping Model for 3-D Visual Tracking With Uncalibrated Eye-in-Hand Robotic System

Neural Networks for Multi-Finger Robot Hand Control

Image Jacobian Matrix Estimation Based on Online Support Vector Regression

Contact Info

Product

Resources

About