Sangrok Jin

Six-Degree-of-Freedom Hovering Control of an Underwater Robotic Platform With Four Tilting Thrusters via Selective Switching Control

We introduce a new underwater robotic platform with a tilting thruster mechanism for hovering motion. The tilting thruster mechanism can implement six-degree-of-freedom (DOF) motion with only four thrusters, but tilting motion makes the system nonlinear. We designed a selective switching controller in order to solve the nonlinear problem, and applied it to the robot system. The selective switching controller divides the six-DOF system into two three-DOF subsystems, and switches between subcontrollers according to the error in real time. Dynamic models of a robotic platform and a disturbance model of an attached manipulator are derived for the control design. Using simulation, the stability condition of control is determined, and the validity of the derived dynamic model of the robotic platform and manipulator is verified through comparison between simulation and experiment. A hovering experiment under a disturbance from the manipulator is performed to verify the robustness of the controller. The experimental results validate the successful hovering ability of the proposed robot.

Optimization of a redundantly actuated 5R symmetrical parallel mechanism based on structural stiffness

A redundantly actuated parallel kinematic machine (PKM) can be used to avoid singularities, normalize manipulability, and increase the stiffness of anon-redundant mechanism. In this study, a redundantly actuated symmetrical PKM with five revolute (5R) joints is optimized for isotropic stiffness in the workspace. The stiffness of the 5R symmetrical PKM is calculated by the superposition of the actuator stiffness and the structural stiffness. We compared the stiffness of anon-redundant PKM and a redundant PKM. Compliance ellipses of the actuator stiffness and the structural stiffness of the non-redundant PKM resulted in the same configurations in the workspace, while those of the redundant PKM resulted in very different configurations. Optimization was performed by determining the optimal actuator torques that are needed to maximize the conditioning index. Optimal results considering structural stiffness can provide a more uniform directional stiffness than optimal results considering the index. When the strength of the linkages in a PKM is weak, the structural stiffness affects the actual stiffness considerably. We believe that the results of this study can be used to help design and control redundant PKMs.

Hovering Performance Improvement by Modifying COG of Underwater Robotic Platform

This paper presents control performance improvement by modifying center of gravity (COG) of an underwater robotic platform. To reduce the oscillation or to increase the positioning accuracy, it is important to accurately know the COG of an underwater robotic platform. The COG is determined by the three measured tilting angles of the platform in different postures. The tilting angle is measured while the platform is hanged by two strings. Using coordinate transformation, the plane of intersection is defined from the angle of the platform and the position of the string. The COG of the robotic platform is directly calculated by the intersected point in three defined planes. The measured COG is implemented to the control algorithm that is pre-designed in the previous research, and the empirical result on tilting gives 48.26% improved oscillation performance comparing to the oscillation result with the ideal COG position.

Hovering underwater robotic platform with four tilting thrusters

This paper represents a new underwater robotic platform with four tilting thrusters for hovering motion. The tilting mechanism can implement the six degrees-of-freedom (DOF) motion with only four thrusters but causes an increase in the nonlinearity of thrust force vector map. In order to overcome the nonlinearity, the selective switching controller is designed, and applied it to the robot system. The selective switching controller chooses each three-DOF sub-controller according to the error in real-time. The experiment, which a robot maintains the position and orientation under a disturbance from an attached manipulator, is performed in a water tank. The results verify the hovering ability of the robot.

Gain Optimization of a Back-Stepping Controller for 6-Dof Underwater Robotic Platform

This paper presents gain optimization of a 6-DOF underwater robotic platform with 4 rotatable thrusters. To stabilize the 6-DOF motion of the underwater robotic platform, a back-stepping controller is designed with 6 proportional gains and 6 derivative gains. The 12 gains of the backstepping controller are optimized to decrease settling time in step response in 6-DOF motion independently. Stability criterion and overshoots are used as a constraint of the optimization problem. Trust-region algorithm and hybrid Taguchi-Random order Coordinate search algorithm are used to determine the optimal parameters, and the results by two methods are analyzed. Additionally, the resulting controller shows improved performance under disturbances.

Theoretical Analyses on Actuator Stiffness and Structural Stiffness of Non-redundant and Redundant Symmetric 5R Parallel Mechanisms

Redundant actuated parallel kinematic machines (PKMs) have been widely researched to increase stiffness of PKMs. This paper presents theoretical analyses on the stiffness of non-redundant and redundant actuated PKM. Stiffness of each mechanism is defined by summation of actuator and structural stiffness; the actuator stiffness is determined from displacements of actuators, and the structural stiffness is determined from deformations of links by external forces. Calculated actuator and structural stiffness of non-redundant PKM show same distribution in entire workspace. On the contrary, the actuator and the structural stiffness of a redundant PKM has very different distribution in the workspace; so, we conclude the structural stiffness of redundant PKM should be considered to design the redundant PKM. The results can be used to design and analyze non-redundant and redundant PKMs.

Optimal design of geometric parameters of a four-bar based manipulator for an underwater robotic platform

In this paper, we propose a starfish capturing four-bar based manipulator for an underwater robotic platform. The manipulator is composed of six links, two torsion springs, and one motor. The model is parameterized to link lengths and spring constants, and we solve kinematics and dynamics to get the trajectory of the end-effector of the manipulator. Capturing stroke, the distance while the end-effector keeps in contact to the ground, is used for the objective function to increase capturing capacity. Six lengths of the links and two torsion spring constants are used for design variables. The reaction force, the moment, and some geometric conditions are considered as inequality constraints in the optimization process. Through the optimization, the capturing stroke of the manipulator is increased by 56% comparing to the initial stroke.

Optimal Design of a Four-bar Linkage Manipulator for Starfish-Capture Robot Platform

In this paper, we propose an optimal design for starfish capturing manipulator module with fourbar linkage mechanism. A tool link with compliance is attached on the four-bar linkage, and the tool repeats detaching starfish from the ground and putting it into the storage box. Since the tool is not rigid and the manipulator is operating underwater, the trajectory of the tool tip is determined by its dynamics as well as kinematics. We analyzed the trajectory of the manipulator tool tip by quasi-static analysis considering both kinematics and dynamics. In optimization, the lengths of each link and the tool stiffness are considered as control variables. To maximize the capturing ability, capturing stroke of the four-bar manipulator trajectory is maximized. Reaction force and reaction moment, and other kinematic constraints were considered as inequality constraints.

Switching PD-based sliding mode control for hovering of a tilting-thruster underwater robot

This paper presents a switching PD-based sliding mode control (PD-SMC) method for the 6-degree-of-freedom (DOF) hovering motion of the underwater robot with tilting thrusters. Four thrusters of robot can be tilted simultaneously in the horizontal and vertical directions, and the 6-DOF motion is achieved by switching between two thruster configurations. Therefore, the tilting speed of thruster becomes the most essential parameter to determine the stability of hovering motion. Even though the previous PD control ensures stable hovering motion within a certain ranges of tilting speed, a PD-SMC is suggested in this paper by combining PD control with sliding mode control in order to achieve acceptable hovering performance even at the much lower tilting speeds. Also, the sign function in the sliding mode control is replaced by a sigmoid function to reduce undesired chattering. Simulations show that while PD control is effective only for tilting duration of 600 ms, the PD-based sliding mode control can guarantee the stable hovering motion of underwater robot even for the tilting duration of up to 1500 ms. Extensive experimental results confirm the hovering performance of the proposed PD-SMC method is much superior to that of PD method for much larger tilting durations.

Back-stepping control design for an underwater robot with tilting thrusters

A hovering control design based on back-stepping method is proposed for a dynamic model of an underwater robot with tilting thrusters. In order to achieve various underwater tasks, a robotic platform must be able to maintain its position and orientation against ocean currents and reaction forces from the manipulators operation. The underwater robot which has four tilting thrusters can carry out six degrees-of-freedom (DOF) motion. A dynamic model is derived for the underwater robot based on hydrodynamic analysis and nonlinear thrust vector mapping. A hovering controller based on a dynamic model is derived by using a back-stepping control method, and disturbance models, such as ocean currents and reaction from the attached manipulator, are designed. Simulations show reasonable results of the control system under disturbance.