Yili Fu

Bio: Yili Fu is an academic researcher from Harbin Institute of Technology. The author has contributed to research in topics: Computer science & Artificial intelligence. The author has an hindex of 13, co-authored 71 publications receiving 748 citations.

Learning Monocular Mesh Recovery of Multiple Body Parts Via Synthesis

Abstract: In this paper, we focus on simultaneously recovering the 3D mesh of multiple body parts from a single RGB image. One of the main challenges is that available datasets with full-body 3D annotations are very limited. This results in poor generalization ability of existing learning-based methods. Existing optimization-based methods iteratively fit the 3D mesh to the 2d pose, which is very time-consuming. To address these limitations, we propose to integrate multiple 3D single-body-part datasets to create a highly diverse whole-body 3D motion space for learning from controllable synthetics. Compared with the learning-based approaches, the proposed method greatly alleviates the reliance on training data. Compared with the optimization-based approaches, the proposed method is a hundred times faster. Our proposed method also outperforms previous state-of-the-art methods on CMU Panoptic dataset.

Design and implementation of a medical robot for celiac minimally invasive surgery

Abstract: This paper presents the design and implementation of a medical robot for celiac minimally invasive surgery. Firstly, the mechanism structure of the robot is briefly introduced which has the feature of simple structure and high safety. Next, kinematics is presented by using the D-H method and the four groups of arm configuration is proposed with the Maximum Distance Criterion method, which provide the surgeons with guidance of arm configuration setting before operation. Then the hardware structure and the software system of the control system are studied and finally the performance of the robot is tested with the prototype experiments. Experiments show that the designed robot has the good “remote center of motion” performance, accomplishes surgical programming successfully and meets the requirements of the minimally invasion surgery.

Ultrasound-guided prostate percutaneous intervention robot system and calibration by informative particle swarm optimization

Abstract: Abstract Applying a robot system in ultrasound-guided percutaneous intervention is an effective approach for prostate cancer diagnosis and treatment. The limited space for robot manipulation restricts structure volume and motion. In this paper, an 8-degree-of-freedom robot system is proposed for ultrasound probe manipulation, needle positioning, and needle insertion. A novel parallel structure is employed in the robot system for space saving, structural rigidity, and collision avoidance. The particle swarm optimization method based on informative value is proposed for kinematic parameter identification to calibrate the parallel structure accurately. The method identifies parameters in the modified kinematic model stepwise according to parameter discernibility. Verification experiments prove that the robot system can realize motions needed in targeting. By applying the calibration method, a reasonable, reliable forward kinematic model is built, and the average errors can be limited to 0.963 and 1.846 mm for insertion point and target point, respectively.

Mix Frame Visual Servo Control Framework for Autonomous Assistive Robotic Arms

Abstract: Assistive robotic arms (ARAs) that provide care to the elderly and people with disabilities, are a significant part of Human-Robot Interaction (HRI). Presently available ARAs provide non-intuitive interfaces such as joysticks for control and thus, lacks the autonomy to perform daily activities. This study proposes that, for inducing autonomous behavior in ARAs, visual sensors integration is vital, and visual servoing in the direct Cartesian control mode is the preferred method. Generally, ARAs are designed in a configuration where its end-effector’s position is defined in the fixed base frame while orientation is expressed in the end-effector frame. We denoted this configuration as ‘mixed frame robotic arms’. Consequently, conventional visual servo controllers which operate in a single frame of reference are incompatible with mixed frame ARAs. Therefore, we propose a mixed-frame visual servo control framework for ARAs. Moreover, we enlightened the task space kinematics of a mixed frame ARAs, which led us to the development of a novel “mixed frame Jacobian matrix”. The proposed framework was validated on a mixed frame JACO-2 7 DoF ARA using an adaptive proportional derivative controller for achieving image-based visual servoing (IBVS), which showed a significant increase of 31% in the convergence rate, outperforming conventional IBVS joint controllers, especially in the outstretched arm positions and near the base frame. Our Results determine the need for the mixed frame controller for deploying visual servo control on modern ARAs, that can inherently cater to the robotic arm’s joint limits, singularities, and self-collision problems.

Wheel leg type humanoid robot with internal oil flowing

Abstract: The invention discloses a wheel leg type humanoid robot with internal oil flowing, relates to the technical field of robots, and aims to solve the problems that when a hydraulic driving unit and a power unit are connected by adopting a rubber hose, the hose is long, the pipe arrangement is complex, and the movement flexibility of a conventional hydraulically-driven robot is influenced. The humanoid robot comprises a head, shoulders, a body, a hip pelvic bone, a first hydraulic connecting rod assembly, a second hydraulic connecting rod assembly, a third hydraulic connecting rod assembly, a hydraulic cylinder, two mechanical arms, two thighs, two shanks and two driving wheels. The head, the shoulders and the body are sequentially and fixedly connected together from top to bottom; the body ishinged to the hip pelvic bone and driven by the hydraulic cylinder, each mechanical arm is hinged to the corresponding shoulder and driven by the first hydraulic connecting rod assembly to swing leftwards and rightwards, and each thigh is hinged to the hip pelvic bone and driven by the second hydraulic connecting rod assembly to swing forwards and backwards. Each joint adopts a hose-free oil way,so that the pipeline abrasion fault is greatly reduced, and the safety and the reliability of the whole machine are improved.

Continuum Robots for Medical Applications: A Survey

Abstract: In this paper, we describe the state of the art in continuum robot manipulators and systems intended for application to interventional medicine. Inspired by biological trunks, tentacles, and snakes, continuum robot designs can traverse confined spaces, manipulate objects in complex environments, and conform to curvilinear paths in space. In addition, many designs offer inherent structural compliance and ease of miniaturization. After decades of pioneering research, a host of designs have now been investigated and have demonstrated capabilities beyond the scope of conventional rigid-link robots. Recently, we have seen increasing efforts aimed at leveraging these qualities to improve the frontiers of minimally invasive surgical interventions. Several concepts have now been commercialized, which are inspiring and enabling a current paradigm shift in surgical approaches toward flexible access routes, e.g., through natural orifices such as the nose. In this paper, we provide an overview of the current state of this field from the perspectives of both robotics science and medical applications. We discuss relevant research in design, modeling, control, and sensing for continuum manipulators, and we highlight how this work is being used to build robotic systems for specific surgical procedures. We provide perspective for the future by discussing current limitations, open questions, and challenges.

Ferromagnetic soft continuum robots

Abstract: Small-scale soft continuum robots capable of active steering and navigation in a remotely controllable manner hold great promise in diverse areas, particularly in medical applications. Existing continuum robots, however, are often limited to millimeter or centimeter scales due to miniaturization challenges inherent in conventional actuation mechanisms, such as pulling mechanical wires, inflating pneumatic or hydraulic chambers, or embedding rigid magnets for manipulation. In addition, the friction experienced by the continuum robots during navigation poses another challenge for their applications. Here, we present a submillimeter-scale, self-lubricating soft continuum robot with omnidirectional steering and navigating capabilities based on magnetic actuation, which are enabled by programming ferromagnetic domains in its soft body while growing hydrogel skin on its surface. The robot's body, composed of a homogeneous continuum of a soft polymer matrix with uniformly dispersed ferromagnetic microparticles, can be miniaturized below a few hundreds of micrometers in diameter, and the hydrogel skin reduces the friction by more than 10 times. We demonstrate the capability of navigating through complex and constrained environments, such as a tortuous cerebrovascular phantom with multiple aneurysms. We further demonstrate additional functionalities, such as steerable laser delivery through a functional core incorporated in the robot's body. Given their compact, self-contained actuation and intuitive manipulation, our ferromagnetic soft continuum robots may open avenues to minimally invasive robotic surgery for previously inaccessible lesions, thereby addressing challenges and unmet needs in healthcare.