Analytic Modeling for Precise Speed Tracking of Multilink Robotic Fish

Verma, Saurab; Xu, Jian‐Xin

doi:10.1109/tie.2017.2779431

Cited by 28 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where ρ represents the density of the fluid, v in represents the relative flow speed as defined in (7), w is the average chord length of the wing and µ is the viscosity of the fluid. In our case, the water temperature was controlled by the water tunnel to be constantly 25…”

Section: Parameter Settingsmentioning

confidence: 99%

Biomimetic Realization of a Robotic Penguin Wing: Design and Thrust Characteristics

Shen

Harada

Katagiri

et al. 2021

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

Flapping flippers or fins are widely employed in biomimetic and bioinspired aquatic robots that imitate natural swimmers such as fish and dolphins. This paper involves robotic biomimetics of the penguin, which is often overlooked as an excellent swimmer benefiting from agile wings. To achieve equivalent swimming skills in an aquatic robot, the wing motion of a real penguin was investigated. Based on the findings, we developed a three-degree-of-freedom (3-DoF) robotic penguin wing that is able to implement the same flapping, feathering and pitch motions as in natural penguins. A kinematic analysis is presented, and the generation of swimming motions is demonstrated. We designed experiments to study the thrust characteristics of the motion of each joint. The results suggest that the flapping motion generates the main thrust during locomotion and the pitch angle can effectively change the thrust direction, while the feathering motion enables active control of the angle of attack (AoA). According to the results, stalling of the wing occurs near an AoA of 10 • in the steady state; however, the feathering motion can avoid the stall and boost the net thrust by a factor of up to 7 relative to the net thrust without active feathering. Furthermore, the introduction of the feathering motion reduces the torque required to actuate the flapping motion. The hydrodynamics of the wing in different flow states are also discussed.

show abstract

Section: Parameter Settingsmentioning

confidence: 99%

Biomimetic Realization of a Robotic Penguin Wing: Design and Thrust Characteristics

Shen

Harada

Katagiri

et al. 2021

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

show abstract

“…However, the non-linear nature of the developed hydrodynamic model hinders end-to-end motion control. Hence, an alternate method is proposed by Verma et al 24 using an analytical model that considers both thrust delay and thrust gradient for non-linear input and output mapping. This work proposes an affine-in-control strategy and SMC for speed tracking of robotic fish.…”

Section: Introductionmentioning

confidence: 99%

Real-time implementation of deep reinforcement learning controller for speed tracking of robotic fish through data-assisted modeling

Duraisamy

Santhanakrishnan

Amirtharajan

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

This article proposes real-time speed tracking of two-link surface swimming robotic fish using a deep reinforcement learning (DRL) controller. Hydrodynamic modelling of robotic fish is done by virtue of Newtonian dynamics and Lighthill’s kinematic model. However, this includes external unsteady reactive forces that cannot be modeled accurately due to the distributed nature of hydrodynamic behavior. Therefore, a novel data-assisted dynamic model and control method is proposed for the speed tracking of robotic fish. Initially, the cruise speed motion data are collected through experiments. The water-resistance coefficient is estimated using the least mean square fit, which is then adopted in the model. Subsequently, a closed-loop discrete-time DRL controller trained through a soft actor-critic (SAC) agent is implemented through simulations. SAC overcomes the brittleness problem encountered by other policy gradient approaches by encouraging the policy network for maximum exploration and not assigning a higher probability to any single part of actions. Due to this robustness in the policy learning, the convergence error becomes low in RL-SAC than RL-DDPG controller. The simulation results verify that the DRL-SAC control with data-assisted modelling substantially improves the speed tracking performance. Further, this controller is validated in real-time, and it is observed that the SAC-trained controller tracks the desired speed more accurately than the DDPG controller.

show abstract

“…Traditional methods need a mathematic model to describe the dynamics of the environment and to choose actions. 2,[16][17][18][19] In general, the model can be built with three main approaches: analytical methods, numerical simulations, and experimental measurements. 1 However, due to the extreme complexity of the hydrodynamics of swimming organisms, it is difficult to establish an accurate mathematic model allowing motion control.…”

Section: Introductionmentioning

confidence: 99%

A numerical simulation of target-directed swimming for a three-link bionic fish with deep reinforcement learning

Zhu

Pang

2022

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

An accurate and robust feedback control system is of great importance for bionic robotic fish to perform complex tasks. This work presents a numerical study of target-directed swimming for a three-link bionic fish with a feedback control system based on deep reinforcement learning (DRL). The simulation is achieved by using a hybrid method of the DRL method and the immersed boundary–lattice Boltzmann method (IB–LBM). This framework makes use of the high computational efficiency of the IB-LBM for the generation of massive motion data needed in the DRL training. The fish is first trained to swim towards a static target from random orientation and distance. The only information available to the fish is its orientation and distance from the target. It learns an accurate and subtle control policy after the training. Then the control policy is applied to a moving target. Even though the fish encounters some situations that never happened in the training, it can choose appropriate actions to follow the target in close proximity and automatically adapt its velocity with the velocity of the target. Those simulations demonstrate the accuracy, robustness of the control method, and its ability to adapt to new situations.

show abstract

Analytic Modeling for Precise Speed Tracking of Multilink Robotic Fish

Cited by 28 publications

References 28 publications

Biomimetic Realization of a Robotic Penguin Wing: Design and Thrust Characteristics

Biomimetic Realization of a Robotic Penguin Wing: Design and Thrust Characteristics

Real-time implementation of deep reinforcement learning controller for speed tracking of robotic fish through data-assisted modeling

A numerical simulation of target-directed swimming for a three-link bionic fish with deep reinforcement learning

Contact Info

Product

Resources

About