<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>When a human gives an order to a robot, the robot must often use its vision to ascertain the human’s indication. In our previous paper, to develop a system where robots precisely receive and obey human orders in daily work spaces, we proposed an experimental system for finger direction recognition (FDR) in 3D space using stereo matching by mounting two cameras on the robot. In this paper, we evaluate this system for FDR in 3D space by performing a series of evaluation experiments using a turntable capable of fixing a hand in a specific finger direction. We estimated various finger directions and distances for two major pointing hand forms (Forms 1 and 2) and evaluated the precision of θ and Φ. We conclude that the θ and Φ estimations are valid because the estimation error is almost within 10 ° when the distance between the camera and object is less than 110 and 80 cm for θ and Φ for Forms 1 and 2, respectively. Finally, we applied our FDR system to the communication between a robot and a person by visual and tactile sensations. In the application test, the robot recognized the object at which the person pointed and put it in the person’s palm.</p></abstract>

When a human gives an order to a robot, the robot must often use its vision to ascertain the human’s indication. In our previous paper, to develop a system where robots precisely receive and obey human orders in daily work spaces, we proposed an experimental system for finger direction recognition (FDR) in 3D space using stereo matching by mounting two cameras on the robot. In this paper, we evaluate this system for FDR in 3D space by performing a series of evaluation experiments using a turntable capable of fixing a hand in a specific finger direction. We estimated various finger directions and distances for two major pointing hand forms (Forms 1 and 2) and evaluated the precision of θ and Φ. We conclude that the θ and Φ estimations are valid because the estimation error is almost within 10 ° when the distance between the camera and object is less than 110 and 80 cm for θ and Φ for Forms 1 and 2, respectively. Finally, we applied our FDR system to the communication between a robot and a person by visual and tactile sensations. In the application test, the robot recognized the object at which the person pointed and put it in the person’s palm.

Evaluation of Finger Direction Recognition Method for Behavior Control of Robot

International Journal on Smart Sensing and Intelligent Systems

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

{"article-title":"Evaluation of Finger Direction Recognition Method for Behavior Control of Robot"}

When a human gives an order to a robot, the robot must often use its vision to ascertain the human’s indication. In our previous paper, to develop a system...

Evaluation of Finger Direction Recognition Method for Behavior Control of Robot

Publié en ligne: 31 déc. 2013

Pages: 2308 - 2333

Reçu: 20 nov. 2013

Accepté: 15 déc. 2013

DOI: https://doi.org/10.21307/ijssis-2017-640

© 2013 T. Ikai et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.