Search for: All records

Deep Reinforcement Learning (DRL) has shown its capability to solve the high degrees of freedom in control and the complex interaction with the object in the multi-finger dexterous in-hand manipulation tasks. Current DRL approaches lack behavior constraints during the learning process, leading to aggressive and unstable policies that are insufficient for safety-critical in-hand manipulation tasks. The centralized learning strategy also limits the flexibility to fine-tune each robot finger's behavior. This work proposes the Finger-specific Multi-agent Shadow Critic Consensus (FMSC) method, which models the in-hand manipulation as a multi-agent collaboration task where each finger is an individual agent and trains the policies for the fingers to achieve a consensus across the critic networks through the Information Sharing (IS) across the neighboring agents and finger-specific stable manipulation objectives based on the state-action occupancy measure, a general utility of DRL that is approximated during the learning process. The methods are evaluated in two in-hand manipulation tasks on the Shadow Hand. The results show that FMSC+IS converges faster in training, achieving a comparable success rate and much better manipulation stability than conventional DRL methods. 

Dexterous telemanipulation is crucial in advancing human-robot systems, especially in tasks requiring precise and safe manipulation. However, it faces significant challenges due to the physical differences between human and robotic hands, the dynamic interaction with objects, and the indirect control and perception of the remote environment. Current approaches predominantly focus on mapping the human hand onto robotic counterparts to replicate motions, which exhibits a critical oversight: it often neglects the physical interaction with objects and relegates the interaction burden to the human to adapt and make laborious adjustments in response to the indirect and counter-intuitive observation of the remote environment. This work develops an End-Effects-Oriented Learning-based Dexterous Telemanipulation (EFOLD) framework to address telemanipulation tasks. EFOLD models telemanipulation as a Markov Game, introducing multiple end-effect features to interpret the human operator’s commands during interaction with objects. These features are used by a Deep Reinforcement Learning policy to control the robot and reproduce such end effects. EFOLD was evaluated with real human subjects and two end-effect extraction methods for controlling a virtual Shadow Robot Hand in telemanipulation tasks. EFOLD achieved real-time control capability with low command following latency (delay<0.11s) and highly accurate tracking (MSE<0.084 rad). 

Wooden utility poles are one of the most commonly used utility carriers in North America. Even though they are given different protection treatments, wooden utility poles are prone to have defects that are mainly caused by temperature, oxygen, moisture, and high potential hydrogen levels after decades of being exposed in open-air areas. In order to meet the growing demand regarding their maintenance and replacement, an effective health evaluation technology for wooden utility poles is essential to ensure normal power supply and safety. However, the commonly used hole-drilling inspection method always causes extra damage to wooden utility poles and the precision of health evaluation highly relies on technician experience at present. Therefore, a non-destructive health evaluation method with frequency-modulated empirical mode decomposition (FM-EMD) and Laplace wavelet correlation filtering based on dynamic responses of wooden utility poles was proposed in this work. Specifically, FM-EMD was used to separate multiple confusing closely-spaced vibration modes due to nonlinear properties of wooden utility poles into several single modes. The instantaneous frequency and damping factor of the decomposed signal of each single mode of the dynamic response of a wooden utility pole could be determined using Laplace wavelet correlation filtering with high precision. The health status of a wooden utility pole could then be estimated according to the extracted instantaneous frequency and damping factor of the decomposed signal of each single mode. The proposed non-destructive health evaluation method for wooden utility poles was tested in the field and achieved successful results.