Researchers from China and the UK have developed a novel deep learning method for autonomous mobile manipulators in unstructured environments, which could facilitate the autonomous operation of underwater robots.
Compared with traditional industrial robots in manufacturing, it is more challenging for an autonomous robot to work safely in dynamic and unstructured environments, such as vast space, open land and the deep sea. Robot autonomy in uncontrolled scenarios requires significantly extra capabilities, including perception, navigation, decision-making and manipulation.
Researchers from the Shenyang Institute of Automation under the Chinese Academy of Sciences and the Edinburgh Centre for Robotics in the UK have constructed a novel deep-learning-based control system to achieve autonomous mobile manipulation in dynamic and unstructured environments.
The system uses a deep learning method to perceive and understand the environment and targets through an on-board camera. Then, it uses the acquired information and the robot state to autonomously control the robot.
Extensive simulation and experiment results show that the proposed mobile manipulation system can grasp different types of objects autonomously in various simulations and real-world scenarios.
The research lays a foundation for the autonomous operation of complex underwater robot systems, according to the team.
The research was published in the journal Sensors. (Xinhua)
The Chinese-developed underwater robot "Tansuo" conducted its maiden test dive in the South China Sea Monday. The robot is 3.5 meters long and 1.5 meters wide and can dive to a depth of 4,500 meters. It will conduct 20 hours of collaborative operations with the unmanned s...
China's most advanced marine science expedition vessel, Kexue, has successfully retrieved its underwater robot from the South China Sea, bringing back 186 gigabytes of high-definition video and data, media reported Monday. According to the Science and Technology Daily, th...
86-10-68597521 (day)
86-10-68597289 (night)
86-10-68511095 (day)
86-10-68512458 (night)
cas_en@cas.cn
52 Sanlihe Rd., Xicheng District,
Beijing, China (100864)
Copyright © 2002 - Chinese Academy of Sciences
