Qinyuan Ren

Locomotion Learning for an Anguilliform Robotic Fish Using Central Pattern Generator Approach

In this paper, we present locomotion learning for an Anguilliform robotic fish using a central pattern generator (CPG) approach. First, we give the overall structure of the CPG. Different from a traditional CPG that contains only coupled oscillators, our CPG consists of coupled Andronov-Hopf oscillators, an artificial neural network (ANN), and an outer amplitude modulator. Coupled oscillators, which possess a limit-cycle character, are used to generate inputs to excite the ANN. The ANN serves as a learning mechanism, from which we can obtain desired waveforms. By inputting different signals to the ANN, different desired locomotion patterns can be obtained. Outer amplitude modulator resizes the amplitudes of the ANN outputs according to task specifications. The CPG possess temporal scalability, spatial scalability, and phase-shift property; thus, we can obtain desired amplitudes, oscillation frequencies, and phase differences by tuning corresponding parameters. By extracting the swimming pattern from a real fish and using the CPG approach, we successfully generate a new swimming pattern and apply it to the robotic fish. The new pattern reserves the swimming characters of the real fish, and it is more suitable to be applied to the robotic fish. By using the new pattern, the robotic fish can perform both forward locomotion and backward locomotion, which are validated by experiments.

Precise Speed Tracking Control of a Robotic Fish Via Iterative Learning Control

In this paper, we present a novel work in which an iterative learning control (ILC) method is applied to a two-link Carangiform robotic fish in real time and achieves precise speed tracking performance. By virtue of the Lagrangian mechanics method, we establish a mathematical model for the robotic fish. The robotic fish model is highly nonlinear and nonaffine in control input, which hinders the applicability of most control methods that require affine-in-input. ILC is suitable because it works for such circumstances. A P-type ILC algorithm is adopted for speed tracking tasks of the robotic fish. The rigorous convergence analysis is derived based on composite energy function (CEF). In practice, the precise model of robotic fish is difficult to be obtained due to many uncertain factors. By employing ILC, the speed tracking control performance can be improved significantly without using the perfect model. Both simulations and experiments are conducted to illustrate the effectiveness of ILC, and excellent speed tracking is achieved for the robotic fish.

A GIM-Based Biomimetic Learning Approach for Motion Generation of a Multi-Joint Robotic Fish

In this paper, we propose a biomimetic learning approach for motion generation of a multi-joint robotic fish. Based on a multi-joint robotic fish model, two basic Carangiform swimming patterns, namely “cruise” and “C sharp turning”, are extracted as training samples from the observations of real fish swimming. A General Internal Model (GIM), which is an imitation of Central Pattern Generator (CPG) in nerve systems, is adopted to learn and to regenerate coordinated fish behaviors. By virtue of the universal function approximation ability and the temporal/spatial scalabilities of GIM, the proposed learning approach is able to generate the same or similar fish swimming patterns by tuning two parameters. The learned swimming patterns are implemented on a multi-joint robotic fish in experiments. The experiment results verify the effectiveness of the biomimetic learning approach in generating and modifying locomotion patterns for the robotic fish.

Locomotion Generation and Motion Library Design for an Anguilliform Robotic Fish

In this paper, modeling, locomotion generation, motion library design and path planning for a real prototype of an Anguilliform robotic fish are presented. The robotic fish consists of four links and three joints, and the driving forces are the torques applied to the joints. Considering kinematic constraints and hydrodynamic forces, Lagrangian formulation is used to obtain the dynamic model of the fish. Using this model, three major locomotion patterns of Anguilliform fish, including forward locomotion, backward locomotion and turning locomotion are investigated. It is found that the fish exhibits different locomotion patterns by giving different reference joint angles, such as adding reversed phase difference, or adding deflections to the original reference angles. The results are validated by both simulations and experiments. Furthermore, the relations among the speed of the fish, angular frequency, undulation amplitude, phase difference, as well as the relationship between the turning radius and deflection angle are investigated. These relations provide an elaborated motion library that can be used for motion planning of the robotic fish.

A Data-driven Motion Control Approach for a Robotic Fish

In this paper we propose a data-driven motion control approach for a biomimetic robotic fish. The task of the motion control is to achieve desired motion by means of controlling the fish-like swimming gaits of the robot. Due to the complexity of hydrodynamics, it is impossible to derive an analytic model that can precisely describe the interaction between the robotic fish and surrounding water during motion. To address the lack of the robotic model, we explore data-based modeling and control design methods. First, through biomimetic learning from real fish motion data, a General Internal Model (GIM) is established. GIM translates fish undulatory body motion into robotic joint movement; associates the fish gait patterns, such as cruise and turning, with corresponding joint coordination; and adjusts the robotic velocity by GIM parameters. Second, by collecting robotic motion data at a set of operating points, we obtain the quantitative mapping from GIM tuning parameters to robotic speed. Third, applying the quantitative mapping and using GIM parameters as manipulating variables, a feedforward control is computed according to the desired speed, which greatly expedites the initial movement of the robotic fish. Fourth, Proportional- Integral-Derivative (PID) is employed as feedback control, together with an inverse mapping that compensates for the nonlinearity appeared in the quantitative mapping. Fifth, modified Iterative Feedback Tuning (IFT) is developed as an appropriate data-driven tuning approach to determine controller gains. By switching between feedforward and feedback, the motion performance is improved. Finally, real-time control of robotic fish is implemented on a two-joint platform, and two representative swimming gaits, namely “cruise” and “cruise in turning”, are achieved.

Generation of robotic fish locomotion through biomimetic learning

This paper presents a novel biomimetic learning approach for a Carangiform robotic fish to learn swimming locomotion. A video recording system is first set up to capture real fish behaviors that are used as the training samples. Three basic Carangiform swimming motion patterns, “cruise”, “cruise in turning” and “C sharp turn”, are extracted from robotic perspective. A general internal model (GIM) is adopted as a universal central pattern generator (CPG). Based on the universal function approximation ability and the temporal/spatial scalabilities of GIM, biomimetic learning is performed such that the robotic fish is able to learn to generate the same or similar fish swimming motion patterns. The three swimming motion patterns are implemented on a multi-joint robotic fish. The effectiveness of the biomimetic learning approach is verified through experiment results.

Motion Control of a multi-joint robotic fish based on biomimetic learning

In this paper, a biomimetic learning approach is applied for motion control of a multi-joint robotic fish. In the learning approach, a general internal model (GIM) is employed to learn coordinated fish-like locomotion from observing live fish swimming. Owing to the scalabilities of the GIM, the learning approach is able to regenerate similar swim patterns with different temporal/spatial scales in the robot. Through experimental analysis, we find out that the motion states, namely speed and orientation, can be controlled by tuning the GIM parameters as well. Based on this control mechanism, feedback control strategies are designed to achieve desired motion. Finally, the effectiveness of the proposed motion control approach is verified by experiments.

Mimicry of fish swimming patterns in a robotic fish

This paper1, presents a novel approach for realizing Carangiform fish swimming patterns by a robotic fish. A video recording system is first set up to capture real fish behaviors. From robotic perspective, three basic Carangiform fish swimming patterns, “cruise”, “cruise in turning”, and “C sharp turning”, are extracted. Base on observations, the mapping between the fish action parts (angular displacements) and the fish swimming patterns are formulated. Hence, the three swimming motion patterns are implemented on a multi-joint robotic fish. Finally, the effectiveness of the approach is verified through experiment results.

A motion control approach for a robotic fish with iterative feedback tuning

This paper proposes a model-free motion control approach for a robotic fish. The fish-like swimming gaits first are generated by a general internal model (GIM)-based learning approach for the robot. Then, a feedforward controller and a proportional-integral-derivative (PID)-based feedback controller are scheduled to control the swimming gaits of the robot to achieve desired motion. To improve the performance of the feedback controller and avoid tedious manual tuning, a pure data-driven iterative feedback (IFT) method is adopted for tuning the parameters of the feedback controller. Finally, experiment results verify the effectiveness of the motion control approach.

Motion control for a multi-joint robotic fish with the pectoral fins assistance

In this paper, we present a novel motion control approach for a multi-joint robotic fish with the pectoral fins assistance. The robotic fish undulates its body and tail to produce the basic swimming patterns that are generated by a general internal model (GIM), while twists its pectoral fins to assist the body to reach the desired direction. The hydrodynamics of the pectoral fin movements is first analyzed. Then the motion control strategies in two basic patterns, namely “cruise” and “C sharp turning”, are explored. Since it is difficult to explicitly formulate the hydrodynamics of the pectoral fin movements, a model-freed fuzzy logic control (FLC) is employed in the robotic fish control system. Finally, the effectiveness of the approach is verified through experiment results.