首页|Researchers from Tsinghua University Discuss Findings in Robotics (Inverse-reinforcement-learning-based Robotic Ultrasound Active Compliance Control In Uncertain Environments)
Researchers from Tsinghua University Discuss Findings in Robotics (Inverse-reinforcement-learning-based Robotic Ultrasound Active Compliance Control In Uncertain Environments)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
IEEE
Data detailed on Robotics have been presented. According to news originating from Beijing, People’s Republic of China, by NewsRx correspondents, research stated, “Robotic ultrasound systems (RUSs) have gained increasing attention because they can automate repetitive procedures and relieve operators’ workloads. However, the complexity and uncertainty of the human surface pose a challenge for stable scanning control.” Financial supporters for this research include National Natural Science Foundation of China (NSFC), China Postdoctoral Science Foundation, National Key Research and Development Program of China, Beijing Natural Science Foundation. Our news journalists obtained a quote from the research from Tsinghua University, “This article proposes a general active compliance control strategy based on inverse reinforcement learning (IRL) to perform adaptable scanning for uncertain and unstructured environments. We analyze the manual scanning process pattern and propose a velocity-and-force-related control strategy to achieve variable force control and handle unpredictable deformation. Then, a hybrid policy optimization framework is proposed to improve transferability. In this framework, a reinforcement learning policy with a predefined reward is built to establish the relationship between contact force and posture. Furthermore, the policy is re-optimized using IRL and generated demonstrations for IRL training. The policy is trained on simple standard phantoms and further evaluated for stability and transferability in unseen and complex environments. Quantitative results show that the difference between the proposed method and the three-dimensional (3-D) reconstructed model in terms of posture is (2.3 +/-1.3 degrees, 1.9 +/-1.2 degrees) in continuous scans.”
BeijingPeople’s Republic of ChinaAsiaEmerging TechnologiesMachine LearningReinforcement LearningRoboticsRobotsTsinghua University
NETL
NSTL
IEEE
