Human-Manipulator Interface Using Particle Filter

Du, Guanglong; Zhang, Ping; Wang, Xueqian

doi:10.1155/2014/692165

Cited by 7 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, Particle Filter (PF) approach is a technique of Bayesian sequential importance sampling. Both work in [25,26] utilized this approach in their studies. While work in [26] defined the human gestures and real time human tracking from depth data using PF method, work in [25] utilized a human -robot interface system which incorporates PF and Adaptive Multi-space Transformation (AMT) to track the pose of the human hand for controlling the robot manipulator.…”

Section: Detection and Trackingmentioning

confidence: 99%

“…Both work in [25,26] utilized this approach in their studies. While work in [26] defined the human gestures and real time human tracking from depth data using PF method, work in [25] utilized a human -robot interface system which incorporates PF and Adaptive Multi-space Transformation (AMT) to track the pose of the human hand for controlling the robot manipulator. PF is used to estimate the translation of the human hand while AMT is used to improve the accuracy and reliability in determining the pose of the robot.…”

Section: Detection and Trackingmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Video Based Action Recognition in Sports

Rahmad¹,

As’ari²,

Ghazali³

et al. 2018

IJEECS

View full text Add to dashboard Cite

<p class="Abstract">Sport performance analysis which is crucial in sport practice is used to improve the performance of athletes during the games. Many studies and investigation have been done in detecting different movements of player for notational analysis using either sensor based or video based modality. Recently, vision based modality has become the research interest due to the vast development of video transmission online. There are tremendous experimental studies have been done using vision based modality in sport but only a few review study has been done previously. Hence, we provide a review study on the video based technique to recognize sport action toward establishing the automated notational analysis system. The paper will be organized into four parts. Firstly, we provide an overview of the current existing technologies of the video based sports intelligence systems. Secondly, we review the framework of action recognition in all fields before we further discuss the implementation of deep learning in vision based modality for sport actions. Finally, the paper summarizes the further trend and research direction in action recognition for sports using video approach. We believed that this review study would be very beneficial in providing a complete overview on video based action recognition in sports.</p>

show abstract

Section: Detection and Trackingmentioning

confidence: 99%

Section: Detection and Trackingmentioning

confidence: 99%

A Survey of Video Based Action Recognition in Sports

Rahmad¹,

As’ari²,

Ghazali³

et al. 2018

IJEECS

View full text Add to dashboard Cite

show abstract

“…The operation time refers to the time used to control one robot manipulator to move close to another manipulator, adjust the orientation and position of the robots, screw bolt and separate them. To increase the success rate of the experiment, a vision-based technology (Du et al, 2014b) was developed to locate the central axis of the EE of another robot by one camera. If the distance or the angles between the central axes of two robots exceed a certain value, which means that the bolt could not screw into the nut, a vibration feedback would immediately call the operator's attention to change the controlling strategy.…”

Section: Environment Of Experimentsmentioning

confidence: 99%

Hybrid filters and feedback mechanism for wearable-based human-manipulator interface

Zhang

Li²,

Du³

2015

Industrial Robot: An International Journal

Self Cite

View full text Add to dashboard Cite

Purpose – This paper aims to develop a wearable-based human-manipulator interface which integrates the interval Kalman filter (IKF), unscented Kalman filter (UKF), over damping method (ODM) and adaptive multispace transformation (AMT) to perform immersive human-manipulator interaction by interacting the natural and continuous motion of the human operator’s hand with the robot manipulator. Design/methodology/approach – The interface requires that a wearable watch is tightly worn on the operator’s hand to track the continuous movements of the operator’s hand. Nevertheless, the measurement errors generated by the sensor error and tracking failure signicantly occur several times, which means that the measurement is not determined with sufficient accuracy. Due to this fact, IKF and UKF are used to compensate for the noisy and incomplete measurements, and ODM is established to eliminate the influence of the error signals like data jitter. Furthermore, to be subject to the inherent perceptive limitations of the human operator and the motor, AMT that focuses on a secondary treatment is also introduced. Findings – Experimental studies on the GOOGOL GRB3016 robot show that such a wearable-based interface that incorporates the feedback mechanism and hybrid filters can operate the robot manipulator more flexibly and advantageously even if the operator is nonprofessional; the feedback mechanism introduced here can successfully assist in improving the performance of the interface. Originality/value – The interface uses one wearable watch to simultaneously track the orientation and position of the operator’s hand; it is not only avoids problems of occlusion, identification and limited operating space, but also realizes a kind of two-way human-manipulator interaction, a feedback mechanism can be triggered in the watch to reflect the system states in real time. Furthermore, the interface gets rid of the synchronization question in posture estimation, as hybrid filters work independently to compensate the noisy measurements respectively.

show abstract

“…The systems in place can be preliminarily classified as vision based and non-vision based tracking systems. Vision based tracking uses cameras [1][2][3][4][5] or other optical devices like the Kinect sensor [6][7][8][9], Leap motion controller [10][11][12], etc., whereas non-vision based systems often use wearable interfaces [13][14][15][16] to estimate the position and orientation of the hand.…”

Section: Introductionmentioning

confidence: 99%

Sensor Fusion of Leap Motion Controller and Flex Sensors Using Kalman Filter for Human Finger Tracking

Ponraj

Ren

2018

IEEE Sensors J.

View full text Add to dashboard Cite

In our daily life, we, human beings use our hands in various ways for most of our day-to-day activities. Tracking the position, orientation and articulation of human hands has a variety of applications including gesture recognition, robotics, medicine and health care, design and manufacturing, art and entertainment across multiple domains. However, it is an equally complex and challenging task due to several factors like higher dimensional data from hand motion, higher speed of operation, self-occlusion, etc. This paper puts forth a novel method for tracking the finger tips of human hand using two distinct sensors and combining their data by sensor fusion technique.

show abstract

Human-Manipulator Interface Using Particle Filter

Cited by 7 publications

References 28 publications

A Survey of Video Based Action Recognition in Sports

A Survey of Video Based Action Recognition in Sports

Hybrid filters and feedback mechanism for wearable-based human-manipulator interface

Sensor Fusion of Leap Motion Controller and Flex Sensors Using Kalman Filter for Human Finger Tracking

Contact Info

Product

Resources

About