TY - GEN
T1 - A miniature stereo vision system for a biomimetic robotic rat
AU - Zou, Mingjie
AU - Shi, Qing
AU - Li, Chang
AU - Ye, Jing
AU - Ma, Mengchao
AU - Gao, Zihang
AU - Huang, Qiang
AU - Fukuda, Toshio
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/7/2
Y1 - 2018/7/2
N2 - Previously we have developed several robotic rats which can mimic the morphology and behavior of laboratory rats in high similarity. However, these previous robots were not endowed with perception devices, resulting in a limited interaction with rats. To address this problem, we developed a miniature stereo vision system for the robot to perceive surrounding environment. The hardware is composed by two tiny cameras (∼12 mm × 4 mm × 4 mm), wireless transmitter and receiver modules and USB video capture cards. With a delicate design, the two cameras can be rightly embedded into the head of the robot. The robot vision shows excellent measurement performance for obtaining depth information and recognizing unknown object. Additionally, our system is capable of SLAM when tested in a complicated environment by using the well-known ORB-SLAM2 algorithm. The experimental results shows that the average tracking time is 0.031s, which is below the inverse of the camera frame rate (0.033s), thus demonstrating a real-time performance.
AB - Previously we have developed several robotic rats which can mimic the morphology and behavior of laboratory rats in high similarity. However, these previous robots were not endowed with perception devices, resulting in a limited interaction with rats. To address this problem, we developed a miniature stereo vision system for the robot to perceive surrounding environment. The hardware is composed by two tiny cameras (∼12 mm × 4 mm × 4 mm), wireless transmitter and receiver modules and USB video capture cards. With a delicate design, the two cameras can be rightly embedded into the head of the robot. The robot vision shows excellent measurement performance for obtaining depth information and recognizing unknown object. Additionally, our system is capable of SLAM when tested in a complicated environment by using the well-known ORB-SLAM2 algorithm. The experimental results shows that the average tracking time is 0.031s, which is below the inverse of the camera frame rate (0.033s), thus demonstrating a real-time performance.
UR - http://www.scopus.com/inward/record.url?scp=85062513433&partnerID=8YFLogxK
U2 - 10.1109/RCAR.2018.8621690
DO - 10.1109/RCAR.2018.8621690
M3 - Conference contribution
AN - SCOPUS:85062513433
T3 - 2018 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2018
SP - 7
EP - 12
BT - 2018 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2018
Y2 - 1 August 2018 through 5 August 2018
ER -