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.

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.

In this paper, an adaptive locomotion control approach for a hexapod robot is proposed. Inspired from biological neuro control systems, a 3D two-layer artificial center pattern generator (CPG) network is adopted to generate the locomotion of the robot. The first layer of the CPG is responsible for generating several basic locomotion patterns and the functional configuration of this layer is determined through kinematics analysis. The second layer of the CPG controls the limb behavior of the robot to adapt to environment change in a specific locomotion pattern. To enable the adaptability of the limb behavior controller, a reinforcement learning (RL)-based approach is employed to tune the CPG parameters. Owing to symmetrical structure of the robot, only two parameters need to be learned iteratively. Thus, the proposed approach can be used in practice. Finally, both simulations and experiments are conducted to verify the effectiveness of the proposed control approach.

Soft actuators have played an indispensable role in generating compliant motions of soft robots. Among the various soft actuators explored for soft robotic applications, dielectric elastomer actuators (DEAs) have caught the eye with their intriguing attributes similar to biological muscles. However, the control challenge of DEAs due to their strong nonlinear behaviors has hindered the development of DEA-based soft robots. To overcome the control challenge, this paper proposes a bioinspired control approach of DEAs. A three-dimensional muscle-like DEA, capable of large forces and giant deformation, is fabricated and adopted as the control platform. To facilitate the controller design, the dynamic model of the DEA is developed through experimental analysis, which takes electromechanical coupling, viscoelastic effects and dynamics uncertainties into consideration. Motivated by the proprioception of the biological muscles, the self-sensing capability of the actuator is explored and exhibits good accuracy. Thus the self-sensing of the actuator is utilized to provide the sensory feedback in the control loop without the need of additional external sensors. Inspired from the role of the cerebellum in motor learning, a cerebellum model articulation nonlinear controller is proposed to compensate the dynamics uncertainties and to provide motion correction. Finally, the effectiveness of the proposed control approach is verified by both the simulation and the experiments.