A5054598974- Soft Robotics and Applications
- Robotic Path Planning Algorithms
- Robot Manipulation and Learning
- Surgical Simulation and Training
- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
- Anatomy and Medical Technology
- Human Pose and Action Recognition
- Augmented Reality Applications
- Teleoperation and Haptic Systems
- Multimodal Machine Learning Applications
- Machine Learning and Algorithms
- Robotic Mechanisms and Dynamics
- Modular Robots and Swarm Intelligence
- Advanced Image and Video Retrieval Techniques
- Reinforcement Learning in Robotics
- Advanced Sensor and Energy Harvesting Materials
- Autonomous Vehicle Technology and Safety
- Robotic Locomotion and Control
- Advanced Neural Network Applications
- Advanced Materials and Mechanics
- Image Processing Techniques and Applications
- Cybercrime and Law Enforcement Studies
- Medical Image Segmentation Techniques
- 3D Shape Modeling and Analysis
University of California, San Diego
2016-2025
University of British Columbia
2010-2025
Peking University
2023
Arizona State University
2021
Jet Propulsion Laboratory
2021
New York University Langone Orthopedic Hospital
2018-2019
Orthopaedic Hospital
2018
University of Southampton
2013-2017
Stanford University
2014-2015
Palo Alto University
2015
Robotic artificial muscles are a subset of that capable producing biologically inspired motions useful for robot systems, i.e., large power-to-weight ratios, inherent compliance, and range motions. These actuators, ranging from shape memory alloys to dielectric elastomers, increasingly popular biomimetic robots as they may operate without using complex linkage designs or other cumbersome mechanisms. Recent achievements in fabrication, modeling, control methods have significantly contributed...
Continuum manipulators offer a means for robot manipulation in constrained environment, where the manipulator body can safely interact with, comply and navigate around obstacles. However, obstacle interactions impose constraints that conform into arbitrary shapes regardless of actuator positions. Generally, these effects cannot be wholly sensed on continuum and, therefore, render model-based controllers incorrect, leading to artificial singularities unstable behavior. We present task-space...
Fast and efficient motion planning algorithms are crucial for many state-of-the-art robotics applications such as self-driving cars. Existing methods become ineffective their computational complexity increases exponentially with the dimensionality of problem. To address this issue, we present Motion Planning Networks (MPNet), a neural network-based novel algorithm. The proposed method encodes given workspaces directly from point cloud measurement generates end-to-end collision-free paths...
Natural muscles exhibit high power-to-weight ratios, inherent compliance and damping, fast actuation dynamic ranges. Unfortunately, traditional robotic actuators have been unable to attain similar properties, especially in a slender muscle-like form factor. Recently, super-coiled polymer (SCP) rejuvenated the promise of an artificial muscle. Constructed from commercial nylon fishing line or sewing thread twisted until coils form, these lightweight shown produce significant mechanical power...
This article describes motion planning networks (MPNet), a computationally efficient, learning-based neural planner for solving problems.MPNet uses to learn general near-optimal heuristics path in seen and unseen environments. It takes environment information such as raw point cloud from depth sensors, well robot's initial desired goal configurations recursively calls itself bidirectionally generate connectable paths. In addition finding directly paths single pass, we show that worst-case...
Traumatic injuries require timely intervention, but medical expertise is not always available at the patient's location. Despite recent advances in telecommunications, surgeons still have limited tools to remotely help inexperienced surgeons. Mixed Reality hints a future where remote collaborators work side-by-side as if co-located; however, we do know how current technology can improve surgical collaboration. Through role-playing and iterative-prototyping, identify collaboration practices...
Artificial intelligence (AI) applications in medical robots are bringing a new era to medicine. Advanced can perform diagnostic and surgical procedures, aid rehabilitation, provide symbiotic prosthetics replace limbs. The technology used these devices, including computer vision, image analysis, haptics, navigation, precise manipulation, machine learning (ML) , could allow autonomous carry out imaging, remote surgery, subtasks, or even entire procedures. Moreover, AI rehabilitation devices...
Robotics have long sought an actuation technology comparable to or as capable biological muscle tissue. Natural muscles exhibit a high power-to-weight ratio, inherent compliance and damping, fast action, dynamic range. They also produce joint displacements forces without the need for gearing additional hardware. Recently, supercoiled commercially available polymer threads (sewing thread nylon fishing lines) been used create significant mechanical power in muscle-like form factor. Heating...
Continuum manipulators are designed to operate in constrained environments that often unknown or unsensed, relying on body compliance conform obstacles. The interaction mechanics between the compliant and environment present significant challenges for traditional robot control techniques based modeling these interactions exactly. In this letter, we describe a hybrid position/force method uses model-less approach. Model-less is recently-described approach learns continuum manipulator Jacobian...
A single thin-film transistor senses, amplifies, and multiplexes force to sense slip close the loop in robotic gripping.
At the beginning of twenty-first century, before power online social networking became apparent, several studies speculated about likely structure organised cybercrime. In light new data on cybercriminal organisations, this paper sets out to revisit their claims. This examines cybercrime by analysing from underground markets previously in operation over Internet. order understand various structures which have manifested, theories are drawn psychology, crime and transaction cost economics...
This article demonstrates that collision detection-intensive applications such as robotic motion planning may be accelerated by performing checks with a machine learning model. We propose Fastron, learning-based algorithm, to model robot's configuration space used proxy detector in place of standard geometric checkers. demonstrate leveraging the results up an order magnitude faster performance robot simulation and than state-of-the-art detection libraries. Our show Fastron learns more 100...
This paper presents a novel miniature uniaxial force sensor for use within beating heart during mitral valve annuloplasty. The measures 5.5 mm in diameter and 12 length provides hollow core to pass instrumentation. A soft elastomer flexure design maintains waterproof seal. Fiber optic transduction eliminates electrical circuitry the heart, acetal components minimize ultrasound-imaging artifacts. Calibration uses nonlinear viscoelastic method, vitro tests demonstrate 0-4-N range with rms...
Sampling-based Motion Planners (SMPs) have become increasingly popular as they provide collision-free path solutions regardless of obstacle geometry in a given environment. However, their computational complexity increases significantly with the dimensionality motion planning problem. Adaptive sampling is one ways to speed up SMPs by particular region configuration space that more likely contain an optimal solution. Although there are wide variety algorithms for adaptive sampling, rely on...
Reinforcement Learning (RL) is a machine learning framework for artificially intelligent systems to solve variety of complex problems. Recent years has seen surge successes solving challenging games and smaller domain problems, including simple though non-specific robotic manipulation grasping tasks. Rapid in RL have come part due the strong collaborative effort by community work on common, open-sourced environment simulators such as OpenAI's Gym that allow expedited development valid...
The supercoiled polymer (SCP) actuator is a recently discovered artificial muscle that demonstrates significant mechanical power, large contraction, and good dynamic range in muscle-like form factor. There has been rapid increase of research efforts devoted to the study SCP actuators. For robotics, actuators overcome specific challenges muscles such as shape memory alloy wires, where limited strain slow dynamics, power consumption had their use. It known hysteresis nonlinearity results from...
We present a method for estimating visual and haptic force feedback on robotic surgical systems that currently do not include significant the operator. Our approach permits to compute contact forces between instruments tissues without additional sensors, relying only endoscopic images acquired by stereoscopic camera. Using an underlying biomechanical model built on-the-fly from organ shape considering tool as boundary conditions acting surface of model, can be estimated at tip tool. At same...
Fast and efficient path generation is critical for robots operating in complex environments. This motion planning problem often performed a robot's actuation or configuration space, where popular pathfinding methods such as A <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> , RRT get exponentially more computationally expensive to execute the dimensionality increases spaces become cluttered complex. On other hand, if one were save entire...
Robotic automation in surgery requires precise tracking of surgical tools and mapping deformable tissue. Previous works on perception frameworks require significant effort developing features for tool tissue tracking. In this work, we overcome the challenge by exploiting deep learning methods perception. We integrated neural networks, capable efficient feature extraction, into processes. By leveraging transfer learning, deep-learning-based approach minimal training data reduced engineering...