scholarly journals Modeling and Control of a Soft Robotic Fish with Integrated Soft Sensing

2021 ◽  
pp. 2000244
Author(s):  
Yu-Hsiang Lin ◽  
Robert Siddall ◽  
Fabian Schwab ◽  
Toshihiko Fukushima ◽  
Hritwick Banerjee ◽  
...  
Keyword(s):  
Robotic Fish ◽  
Soft Sensing ◽  
Modeling And Control ◽  
And Control

Novel Method for the Modeling and Control Investigation of Efficient Swimming for Robotic Fish

2012 ◽  
Vol 59 (8) ◽  
pp. 3176-3188 ◽  
Author(s):  
Li Wen ◽  
Tianmiao Wang ◽  
Guanhao Wu ◽  
Jianhong Liang ◽  
Chaolei Wang
Keyword(s):  
Robotic Fish ◽  
Modeling And Control ◽  
Novel Method ◽  
And Control

MODELING AND CONTROL DESIGN OF AN ANGUILLIFORM ROBOTIC FISH

2012 ◽  
Vol 03 (04) ◽  
pp. 1250018 ◽  
Author(s):  
JIAN-XIN XU ◽  
XUE-LEI NIU ◽  
QIN-YUAN REN
Keyword(s):  
Sliding Mode Control ◽  
Sliding Mode ◽  
Control Design ◽  
Robotic Fish ◽  
Torque Control ◽  
Underactuated System ◽  
Parameter Uncertainties ◽  
Modeling And Control ◽  
Mode Control ◽  
And Control

In this paper, the modeling and control design of a biomimetic robotic fish is presented. The Anguilliform robotic fish consists of N links and N - 1 joints, and the driving forces are the torques applied to the joints. Considering kinematic constraints, Lagrangian formulation is used to obtain the dynamics of the fish model. The computed torque control method is applied first, which can provide satisfactory tracking responses for fish joints. Since this robotic fish is essentially an underactuated system, the reference trajectories for the orientation of the N links are planned in such a way that, at a neighborhood of the equilibrium point, the tracking task of N angles can be achieved by using N - 1 joint torques. To deal with parameter uncertainties that exist in the actual environment, sliding mode control is adopted. Considering feasibility and complexity issues, a simplified sliding mode control algorithm is given. A four-link robotic fish is modeled and simulated, and the results validate the effectiveness of reference planning and the proposed controllers.

Bio-Inspired Design, Modeling, and Control of Robotic Fish Propelled by a Double-Slider-Crank Mechanism Driven Tail

2021 ◽  
Author(s):  
Wenyu Zuo ◽  
Frank Fish ◽  
Zheng Chen
Keyword(s):  
Linear Model ◽  
Swimming Speed ◽  
Robotic Fish ◽  
State Feedback Control ◽  
Modeling And Control ◽  
Swimming Motion ◽  
Heading Angle ◽  
Undulatory Locomotion ◽  
Oscillating Foil ◽  
And Control

Abstract This paper presents the design, modeling, and control of a three-joint robotic fish propelled by a Double-Slider-Crank (DSC) driven caudal fin. DSC is a mechanism that can use one DC motor to achieve oscillating foil propulsion. Its design is guided by a traveling wave equation that mimics a fish’s undulatory locomotion. After multiple tests, the robotic fish displayed good performance in mimicking a real fish’s swimming motion. DSC mechanism is proven to be an effective propulsion technique for a robotic fish. With the help of another servo motor at the first joint of the fish’s tail, the robotic fish can have a two-dimensional free-swimming capability. In experiments, the robotic fish can achieve a swimming speed of 0.35 m/s at 3 Hz, equivalent to 0.98 body length (BL) per second. Its steering rate is proportional to a bias angle. The DSC benefits the control of the robotic fish by independently adjusting its steering and swimming speed. This characteristic is studied in a hydrodynamic model that derives the thrust within a DSC frame. Besides the dynamic model, a semi-physics-based and data-driven linear model is established to connect bias angle to robotic fish’s heading angle. The linear model is used for designing a state feedback control, and the controller has been examined in simulation and experiments.

Learning Based Speed Control of Soft Robotic Fish

2018 ◽  
Author(s):  
Sunil Kumar Rajendran ◽  
Feitian Zhang
Keyword(s):  
Design Method ◽  
Control Design ◽  
Artificial Muscle ◽  
Speed Control ◽  
Robotic Fish ◽  
Design Approach ◽  
Modeling And Control ◽  
Model Free ◽  
Bioinspired Robotics ◽  
And Control

Bioinspired robotics takes advantage of biological systems in nature for morphology, action and perception to build advanced robots of compelling performance and wide application. This paper focuses on the design, modeling and control of a bioinspired robotic fish. The design utilizes a recently-developed artificial muscle named super coiled polymer for actuation and a soft material (silicone rubber) for building the robot body. The paper proposes a learning based speed control design approach for bioinspired robotic fish using model-free reinforcement learning. Based on a mathematically tractable dynamic model derived by approximating the robotic fish with a three-link robot, speed control simulation is conducted to demonstrate and validate the control design method. Exampled with a three-link reduced-order dynamic system, the proposed learning based control design approach is applicable to many and various complicated bioinspired robotic systems.

Modeling and Control for Plant Dynamics Based on Reinforcement Learning

2009 ◽  
Vol 129 (4) ◽  
pp. 363-367
Author(s):  
Tomoyuki Maeda ◽  
Makishi Nakayama ◽  
Hiroshi Narazaki ◽  
Akira Kitamura
Keyword(s):  
Reinforcement Learning ◽  
Modeling And Control ◽  
Plant Dynamics ◽  
And Control
Sign in / Sign up

Export Citation Format

Share Document