Houcheng Li

TL;DR: In this paper , a dynamic system approach is proposed to learn point-to-point motions while keeping the stability of the dynamic system, which is grounded on a Learning from Demonstration (LfD) method based on a neural network, which gets better reproduction performance while guaranteeing the generalization ability.

TL;DR: Wang et al. as mentioned in this paper proposed a passive model predictive impedance control method including two control loops, in which the bottom-loop is driven by a variable impedance controller to achieve the desired compliant interaction behavior, and the top-loop of the proposed controller is used to ensure that the robot states satisfy the passivity constraint by calculating a complementary torque to limit the stored energy of the robot.

TL;DR: A hybrid controller for musculoskeletal robots targeting lifting tasks in industrial metaverse is proposed. The controller utilizes muscle-synergy-based RBF network and adaptive dynamic programming for feedforward and feedback control, respectively. The controller is verified on an upper limb musculoskeletal system, demonstrating its effectiveness for lifting tasks.

TL;DR: This research introduces an innovative methodology aimed at resolving the necessity for physical demonstrations and the explicit designation of starting and ending points for replication, effectively obviating the necessity for physical demonstrations and the explicit designation of starting and ending points for replication.