- Soft Robotics and Applications
- Control and Dynamics of Mobile Robots
- Robotic Path Planning Algorithms
- Robotic Locomotion and Control
- Power Line Inspection Robots
- Micro and Nano Robotics
- Teleoperation and Haptic Systems
- Advanced Algorithms and Applications
- Robot Manipulation and Learning
- Advanced Sensor and Control Systems
- IoT-based Smart Home Systems
- Robotics and Sensor-Based Localization
- Industrial Technology and Control Systems
- Structural Response to Dynamic Loads
- Artificial Immune Systems Applications
- T-cell and B-cell Immunology
- Adhesion, Friction, and Surface Interactions
- Adaptive Dynamic Programming Control
- Anatomy and Medical Technology
- Vehicle Dynamics and Control Systems
- Industrial Automation and Control Systems
- Prosthetics and Rehabilitation Robotics
- Fiber-reinforced polymer composites
- vaccines and immunoinformatics approaches
- Mechanical Behavior of Composites
Beijing Institute of Technology
2009-2024
Guangxi University of Science and Technology
2017-2023
Marymount University
2020
Yokohama National University
2020
Toshiba (Japan)
2020
Helsinki Institute for Information Technology
2005
Harbin Institute of Technology
2003
In this paper we introduce wall-climbing robots of two different absorption schemes: one with single suction cup and the other permanent magnetic crawlers. The former adopts an omnidirectional vehicle can perform remote-control inspection nuclear storage tanks. Moreover, it is further developed for cleaning both ceramic tile glass surfaces high-rise buildings. latter consists types robots: maintenance automation tanks in petrochemical enterprises, which operations sand-blasting,...
Inspired by natural immune systems, artificial systems (AIS) are an emerging kind of computational intelligence paradigm. During the past decade, AIS have gained great research interest in wide engineering fields. Artificial optimization (AIO) methods important partner AIS. They been successfully applied to deal with numerous challenging problems superior performance over classical techniques. This paper gives a concise survey on recent progresses theory as well applications AIO schemes,...
This article introduces a wall-climbing robot that uses the reverse thrust of propeller as adsorption force. The is symmetrically distributed in structure and force before after so it can adapt to surface variety different media materials achieve stable movement wall surfaces. mainly aircraft adsorption. relies on four wheels move forward. forward power comes from combined action component front wheel driving During robot, steering realized by differential control. In this paper, we design...
Tracking control is an essential capability for nonholonomic wheeled mobile robots (NWMR) to achieve autonomous navigation. This paper presents a novel hybrid strategy combined mode-based and actor-critic based deep reinforcement learning method. Based on the Lyapunov method, kinematics law named given obtained with pose errors. Then, tracking problem converted finite Markov decision process, in which defined state contains current errors, inputs one-step After training deterministic policy...
This paper presents a floor cleaning robot equipped with Swedish wheels. It can be used in crowded places such as houses, train station, airport etc. the perform its work autonomous and teleoperated mode. Moreover pivot around without turning, avoid obstacles is provided automatic power management ability. And meanwhile, kinematics for control controlling methods are studied demonstrated. new structure, smooth locomotion capability high working efficiency verified by experimentation.
The wall climbing robot has characteristics of moving smartly and suction reliably on the wall, therefore, this paper proposes a method called critical suction, based negative thrust force. Then mechanics, robot's mechanism performance air-sealed are analyzed, is discussed in theory. Furthermore, fluid model network dynamic system set up. According to response equation pressure cup, process simulated. Finally, effect that key parameters cup have characteristic analyzed conclusions presented.
Coal mine is a dangerous field, various fatal factors menace people. The rescuers have known the tunnel actual situation when disaster in coal occurs. Especially, it very to go into without detect advanced because second explosion may occur at any time. Aiming task, rescue robot can be sent environment and gas content temperature, etc. Meanwhile environmental data are rescuers. This paper introduces robot. It composed of mechanism, control system, communication system sensors, run...
Purpose The purpose of this paper is to present an omni‐directional floor‐cleaning robot equipped with four wheels. research purposes are design a for cleaning jobs in domestic, narrow and crowded places provide robotics‐study platform laboratory. Design/methodology/approach system using Swedish wheels, one dust collector (brush) switching device sort air‐bag sensing designed. kinematics the motion control conditions analyzed. Specifically, method wheels described. Findings configuration...
This paper presents a wall-climbing robot which can stably hold and move on the ground-wall surface. The uses propeller reverse thrust as an adsorption force adapt to surface of several media materials. influence robot’s structural parameters its power system is analyzed by comparing single test prototype test. A analysis performed under two specific situations; when he in transition from ground small slope, slope. state then obtained different conditions. Experimental results show that...
This paper proposes a double propeller wall-climbing robot (DP-Climb) with hybrid adhesion system based on the biomimetic design principle to address problems of single adhesion-powered wall climbing robots (WCRs). Such include poor maneuverability and adaptability orthogonal working surfaces different roughness flatness, weak flexibility ground-wall transition motion, easy stand stilling transition,. Based clinging characteristics creatures, combines rotor units’ reverse thrust, drive...
Coal mine is a dangerous field, varies fatal factors menace people. Especially after coal disaster, rescuer doesn't know tunnel situation. It very to go into without environment detect because second explosion may occur. and rescue robot kind of mobile robot. can gas content, temperature, etc. The date send control man in safe field. This paper designs has many characters suitable tunnel. compose mechanic, electric, computer, control, communicate, sensor, run environment, climb ruins, check...
This paper introduces a centrifugal impeller-based wall climbing robot "BIT Climber". First, by statics and dynamics analyses, mechanical parameter optimization directions for general robots are presented. A hydromechanical model suitable all robots' suction system is built, which important device design. Afterwards the climber's basic configurations of electronical structure described. Finally, experimental results given.
This paper presents a new taxonomy for wall climbing robots, among which 1-NP and NPT methods with continuous rapid locomotion are studied emphatically. It analyses in depth operating principles of fluid machineries that the two use, points out two's difference structure as well merits drawbacks performance. Then it builds dynamic models elements method's suction closed-loop control system, comes up algorithm, shows simulation results discussion. Finally conclusion is given future direction...
A boiler maintenance robot which is equipped many kinds of operation devices proposed. This has compact size and can be used to do works ash cleaning, slag purging, thickness measurement symbol marking for water-cooling tubes in one work cycle. The move smoothly on the tube-wall using double magnetic tracked mechanism designed specially. In this paper, structure robot, method devices, self-protection mechanisms control system are described detail idea design given. Furthermore, through...
For the motion control problem of non-holonomic constrained mobile robots, a point stabilization kinematic law for robot based on deep reinforcement learning is proposed. Firstly, model constructed to build memory learning, including current state robot, action, reward and next which generated through connection between environment. Then, value network parameters in real-time are updated by loss function, composed state-action moment came from target value, network. Next, policy according...
We present a twin-propelled wall-climbing robot that can stably adsorb and move quickly on vertical wall by exploiting the propeller’s reverse thrust as adsorption force wheel movement method. This paper derives theoretically analyses aerodynamic characteristics of mobile system momentum inflow model blade element theory measures single propeller through real-world experiments. Additionally, Computational Fluid Dynamics simulation software analyzes influence structural parameters, such...
Lung cancer is the leading cause of deaths worldwide. Although several lung diagnostic methods are available for nodule biopsy, there limitations in terms accuracy, safety, and invasiveness. Transbronchial needle aspiration (TBNA) a common method diagnosing treating that involves robot-assisted medical flexible moving along curved three-dimensional trajectory, avoiding anatomical barriers to achieve clinically meaningful goals humans. Inspired by puncture angle between tip vessel...
Coal mine detect robot is a kind of mobile robot. It goes into explosion environment and detects gas content, temperature, etc. Environment dates are send to control people in safe field. dangerous field, varies fatal factors menace people. Especially when coal disaster occur. Rescuer doesn't know tunnel actual situation. very go without because second may occur at any time. This paper introduces electric system. has many characters suitable tunnel. compose electric, computer, control,...
With the development of industry and progress society, a kind vehicle that can travel on ground like car fly in air an airplane has become research focus both at home abroad.The method bionic design aerodynamic based combination aesthetic principles, through structure four rotor aircraft, aircraft take-off mode is very suitable for flying takeoff: VTOL flight inspiration, with rotors taking off; optimization method, streamlined style.Using 3D software to build model, beautiful designed meet...
Wall robots must have adsorption and movement functions, the safety of system is a key index which determines robot's performance. The redundant control theory was applied in vacuum absorption wall cleaning robot, its feasibility analyzed, then model established. A double bumps put forward to enhance safety. It's proved theoretically that system, mean time between failures robot extended. experimental data shows practicable adjustable, two pumps can switch rapidly stably.
Conventional vascular interventional surgery (VIS) is performed under fluoroscopic X-ray guidance. Catheter operating systems have been developed to reduce physical stress and irradiation. However, the catheter are all attached surgical bed or fixed at tip of passive manipulator, which inconvenient position adjust system Furthermore, traditional control can't satisfy clinical requirements radiation protecting. Therefore, a novel Active Supporting Medical Manipulator (ASMM) was accurately...
This paper introduced a centrifugal impeller-based wall-climbing robot with the μCOS-II System. Firstly, climber's basic configurations of mechanical were described. Secondly, mechanic analyses walking mechanism was presented, which essential to suction device design. Thirdly, control system including PC remote and STM32 master slave designed. Finally, an experiment conducted test performance negative pressure generating general abilities robot.
A trajectory tracking control algorithm based on deep reinforcement learning is proposed in this paper. It could solve the problem of wheeled mobile robot with nonholonomic constraints. Firstly, by analyzing constraint characteristics robot, kinematics model, dynamics model and motor drive are established. Then, according to a designed using deterministic policy gradient (DDPG) algorithm. Finally, agent trained circle method. The simulation results show that our effectively track target...