This paper proposes a deep reinforcement learning (DRL) method that combines random network distillation (RND) and long short-term memory (LSTM) to address the tracking control problem, while leveraging the inherent symmetry in robotic arm movements to eliminate the need for learning or knowing the system’s dynamic model. In general, the complexity and strong coupling of robotic manipulators make trajectory tracking extremely challenging. Firstly, the prediction network and fixed network are jointly trained using the RND method. The difference in output values between the two networks acts as an internal reward for the robotic manipulator environment. This internal reward mechanism encourages the robotic arm agent to actively explore unpredictable and unknown environmental states, thereby consequently boosting the performance and efficiency of the tracking control for the robotic manipulator. Then, the Soft Actor-Critic (SAC) algorithm, the LSTM network, and the attention mechanism are integrated to resolve the instability problem during training and acquire a stable policy. The LSTM model effectively captures the symmetry and temporal changes in joint angles, while the attention mechanism dynamically prioritizes important features, thereby reducing the instability of the robotic manipulator during tracking tasks and enhancing feature extraction efficiency. The simulation outcomes demonstrate that the proposed method effectively performs the robot tracking task, confirming the efficacy and efficiency of the DRL algorithm.

In this paper, a deep reinforcement learning (DRL) approach based on generative adversarial imitation learning (GAIL) and long short-term memory (LSTM) is proposed to resolve tracking control problems for robotic manipulators with saturation constraints and random disturbances, without learning the dynamic and kinematic model of the manipulator. Specifically, it limits the torque and joint angle to a certain range. Firstly, in order to cope with the instability problem during training and obtain a stability policy, soft actor–critic (SAC) and LSTM are combined. The changing trends of joint position over time are more comprehensively captured and understood by employing an LSTM architecture designed for robotic manipulator systems, thereby reducing instability during the training of robotic manipulators for tracking control tasks. Secondly, the obtained policy by SAC-LSTM is used as expert data for GAIL to learn a better control policy. This SAC-LSTM-GAIL (SL-GAIL) algorithm does not need to spend time exploring unknown environments and directly learns the control strategy from stable expert data. Finally, it is demonstrated by the simulation results that the end effector of the robot tracking task is effectively accomplished by the proposed SL-GAIL algorithm, and more superior stability is exhibited in a test environment with interference compared with other algorithms.

An optimal composite nonlinear feedback control method with integral sliding mode is presented in this paper. The controller extends the travel range of the micro-electromechanical system capacitive micromachined ultrasonic transducer (CMUT). Moreover, enhanced transient response and precise tracking performance is achieved. It is known that CMUT is inherently unstable which results in pull-in phenomenon and it is very sensitive to small perturbations, so one of the major problems is to stabilize the CMUT beyond the pull-in limit with the external disturbances. In addition, the input saturation problem is significant to CMUT. Based on that, a robust control scheme is derived using composite nonlinear feedback control law combined with integral sliding mode control law. Then all the tuning parameters for the proposed control method are converted into a minimization problem and solved by particle swarm optimization algorithm automatically. We verified the effectiveness through extending the travel range of the CMUT gap by three control methods which are proportional integral derivative, composite nonlinear feedback and the method we proposed. The stability and small range tracking performance with three control methods is compared on the pull-in position of CMUT. The simulations show that the proposed control method has desired tracking performance and robustness to external disturbance with input saturation.

In this study, tracking control and saturation nonlinearities problems of a flexible mobile vehicle-manipulator with disturbance uncertainty are investigated. Based on differential equations model, a novel iterative learning control (ILC) algorithm is designed, where two saturation functions and two hyperbolic tangent functions are developed to deal with actuator saturation constraint and an adaptive law is developed to reduce the impact of disturbance uncertainty. With the ILC algorithm, the target of tracking control and vibration attenuation of mobile manipulator can come true. Simulations are executed to verify the effectiveness of designed algorithm.

This paper investigates the stabilization and trajectory tracking problem of wheeled mobile robot with a ceiling-mounted camera in complex environment. First, an adaptive visual servoing controller is proposed based on the uncalibrated kinematic model due to the complex operation environment. Then, an adaptive controller is derived to provide a solution of uncertain dynamic control for a wheeled mobile robot subject to parametric uncertainties. Furthermore, the proposed controllers can be applied to a more general situation where the parallelism requirement between the image plane and operation plane is no more needed. The overparameterization of regressor matrices is avoided by exploring the structure of the camera-robot system, and thus, the computational complexity of the controller can be simplified. The Lyapunov method is employed to testify the stability of a closed-loop system. Finally, simulation results are presented to demonstrate the performance of the suggested control.