High-performance robotic muscles from conductive nylon sewing thread
- Michael C. Yip, G. Niemeyer
- EngineeringIEEE International Conference on Robotics and…
- 26 May 2015
This paper develops a thermomechanical and thermoelectric model of super-coiled polymer actuators, and examines their controllability, and uses them in a robotic hand to demonstrate their applicability as a low-cost, high performance robotic muscle.
On the Control and Properties of Supercoiled Polymer Artificial Muscles
- Michael C. Yip, G. Niemeyer
- EngineeringIEEE Transactions on robotics
- 24 February 2017
This paper describes the working principle of supercoiled polymer (SCP) actuation and explores the controllability and properties of these threads, showing that under appropriate environmental conditions, the threads are suitable as a building block for a controllable artificial muscle.
Motion Planning Networks
- A. H. Qureshi, M. J. Bency, Michael C. Yip
- Computer ScienceIEEE International Conference on Robotics and…
- 14 June 2018
This work presents Motion Planning Networks (MPNet), a neural network-based novel planning algorithm that encodes the given workspaces directly from a point cloud measurement and generates the end-to-end collision-free paths for the given start and goal configurations.
Model-Less Feedback Control of Continuum Manipulators in Constrained Environments
- Michael C. Yip, D. Camarillo
- EngineeringIEEE Transactions on robotics
- 20 March 2014
This work uses an optimal control strategy on a tendon-driven robot to demonstrate model-less control, which allows the manipulator to interact with several constrained environments in a stable manner and is the first work in controlling continuum manipulators without using a model.
Motion Planning Networks: Bridging the Gap Between Learning-Based and Classical Motion Planners
- A. H. Qureshi, Yinglong Miao, A. Simeonov, Michael C. Yip
- Computer ScienceIEEE Transactions on robotics
- 13 July 2019
This article describes motion planning networks (MPNet), a computationally efficient, learning-based neural planner for solving motion planning problems, and shows that worst-case theoretical guarantees can be proven if this neural network strategy is merged with classical sample-based planners in a hybrid approach.
Adversarial Imitation via Variational Inverse Reinforcement Learning
- A. H. Qureshi, Michael C. Yip
- Computer ScienceInternational Conference on Learning…
- 17 September 2018
The results show that the proposed empowerment-regularized maximum-entropy inverse reinforcement learning method not only learns near-optimal rewards and policies that are matching expert behavior but also performs significantly better than state-of-the-art inverse reinforcementlearning algorithms.
Learning-Based Proxy Collision Detection for Robot Motion Planning Applications
- Nikhil Das, Michael C. Yip
- Computer ScienceIEEE Transactions on robotics
- 21 February 2019
This article proposes Fastron, a learning-based algorithm, to model a robot's configuration space to be used as a proxy collision detector in place of standard geometric collision checkers, and shows its application toward autonomous surgical assistance task in shared environments with human-controlled manipulators.
Model-Less Hybrid Position/Force Control: A Minimalist Approach for Continuum Manipulators in Unknown, Constrained Environments
- Michael C. Yip, D. Camarillo
- Computer ScienceIEEE Robotics and Automation Letters
- 5 February 2016
A hybrid control approach for simultaneously controlling end-effector position and forces while still maintaining the minimalist approach of model-less control is described, which can safely and effectively interact with the environment, even when these interactions are arbitrary and unknown constraints.
Modeling and Inverse Compensation of Hysteresis in Supercoiled Polymer Artificial Muscles
- Jun Zhang, K. Iyer, A. Simeonov, Michael C. Yip
- EngineeringIEEE Robotics and Automation Letters
- 11 January 2017
Three new models are formulated to characterize the hysteresis relationship between three coupled variables (voltage input, strain, and load) of an SCP actuator, namely the augmented generalized Prandtl–Ishlinskii model, the augmented Preisach model, and the augmented linear model.
Tissue Tracking and Registration for Image-Guided Surgery
- Michael C. Yip, D. Lowe, S. Salcudean, R. Rohling, C. Nguan
- Computer ScienceIEEE Transactions on Medical Imaging
- 9 August 2012
An integrated framework for accurately tracking tissue in surgical stereo-cameras at real-time speeds is presented and the salient feature framework is extended to support region tracking in order to maintain the spatial correspondence of a tracked region of tissue or a medical image registration to the surrounding tissue.
...
...