scispace - formally typeset
Search or ask a question

Showing papers by "Yili Fu published in 2016"


Proceedings ArticleDOI
01 Dec 2016
TL;DR: The design applied a asymmetrical structure mimicking snake spines and muscles and it has the same level of dexterity compared with rotating joint and the workspace is analyzed via screw theory and vector position equation.
Abstract: This paper proposes a new joint mechanism for a snake-like robot to travel in hazardous areas for rescue tasks. In a complex environment, it is necessary to preform an efficient three-dimensional motion. But, a serial snake-like robot could hardly do that because of the limitation of its 1-DOF joint structure. To meet the requirement of flexible locomotion, a parallel mechanism was proposed which was derived from the biological characteristic of snakes. The design applied a asymmetrical structure mimicking snake spines and muscles and it has the same level of dexterity compared with rotating joint. We obtain control equations for yawing and pitching angle by inverse kinematic solution. Then, the workspace is analyzed via screw theory and vector position equation. A non-model based control serpentine gait and a multiple beat rectilinear motion of snake-like robot with parallel mechanisms are introduced.

6 citations


Proceedings ArticleDOI
01 Aug 2016
TL;DR: The experiment results prove the effectiveness of proposed control method and also present that the designed hydraulic SEA could realize virtual spring leg with stiffness adjustable from 5KN/m to 20 KN/m, which is the stiffness needed in human's normal walking.
Abstract: SLIP model with compliance describes human locomotion very well. A Legged robot's variable stiffness joint could produce an equivalent virtual spring leg, which is the foundation for SLIP model's physical implementation. This paper presents a single leg with knee joint actuated by hydraulic SEA and proposes a variable stiffness control method that is applied to the virtual spring leg. The experiment results prove the effectiveness of proposed control method and also present that the designed hydraulic SEA could realize virtual spring leg with stiffness adjustable from 5KN/m to 20KN/m, which is the stiffness needed in human's normal walking. The robot's single leg weighs about 6kg and could burden at least 60kg load, which provides a way to realize the heavy load legged robot with variable stiffness joint.

5 citations


Proceedings ArticleDOI
27 Jul 2016
TL;DR: In this article, a modular bionic parallel joint mechanism (BPJM) is proposed for the rescue snake robot, which is a key factor for a snake robot adjusting its postures adapted to clutter environments in search and rescue tasks.
Abstract: Joint mechanism is a key factor for a snake robot adjusting its postures adapted to clutter environments in search and rescue tasks. Most joint mechanisms in prior research simply consist of serially connected revolute joints, which are lack of great load carrying ability. Imitating nature snake structure, a modular bionic parallel joint mechanism (BPJM) is proposed for the rescue snake robot. Dynamics analysis of the BPJM is necessary for its optimal design and control, providing the force and constraint that must be resisted by joints, links and actuators. To reduce the dynamics computation load, Newton equation and Euler equation are combined by synchronizing the inertial force and inertial moment with the aid of screw theory. Therefore, the dynamics equations for moving platform and links are formulated in a simplified form. Viscous friction at the joints and external force acting on BPJM, which actually affect the motion, are both considered in the formulation. Finally, the virtual prototype is provided in order to visualize the joint mechanism and the numerical results from the dynamics analysis are given.

2 citations


Proceedings ArticleDOI
01 Dec 2016
TL;DR: Simulation experiments show that the proposed control method can improve the position tracking accuracy of the flexible-joint manipulator and restrain the flexible -joint vibration obviously, and thus verifies that the designed controller can meet the requirements of accurate tracking control of the surgical robot.
Abstract: Due to the surgical robot flexible joint state vectors are not all measurable, in order to reduce the effect of the modeling errors and nonlinear factors of the flexible-joint manipulator caused by gravity and friction, an extended high-gain observer (EHGO) based adaptive control algorithm is designed to perform the position tracking control of the flexible-joint manipulator. First, gravity and friction are fed forward to reduce the nonlinearity of the robotic manipulator. Secondly, an extended high gain observer is used to observe the angular velocity of the joint and correct compensation errors of gravity and friction in real time. Finally, EHGO based adaptive controller is used to eliminate the inertia parameter measurement errors of the dynamic model, and to achieve high precision position tracking control. Simulation experiments show that the proposed control method can improve the position tracking accuracy of the flexible-joint manipulator and restrain the flexible-joint vibration obviously, and thus verifies that the designed controller can meet the requirements of accurate tracking control of the surgical robot.

2 citations