Robust State/Output-Feedback Control of Coaxial-Rotor MAVs Based on Adaptive NN Approach

Design and Experimental Validation of Deep Reinforcement Learning-Based Fast Trajectory Planning and Control for Mobile Robot in Unknown Environment

This article is concerned with the problem of planning optimal maneuver trajectories and guiding the mobile robot toward target positions in uncertain environments for exploration purposes. A hierarchical deep learning-based control framework is proposed which consists of an upper level motion planning layer and a lower level waypoint tracking layer. In the motion planning phase, a recurrent deep neural network (RDNN)-based algorithm is adopted to predict the optimal maneuver profiles for the mobile robot. This approach is built upon a recently proposed idea of using deep neural networks (DNNs) to approximate the optimal motion trajectories, which has been validated that a fast approximation performance can be achieved. To further enhance the network prediction performance, a recurrent network model capable of fully exploiting the inherent relationship between preoptimized system state and control pairs is advocated. In the lower level, a deep reinforcement learning (DRL)-based collision-free control algorithm is established to achieve the waypoint tracking task in an uncertain environment (e.g., the existence of unexpected obstacles). Since this approach allows the control policy to directly learn from human demonstration data, the time required by the training process can be significantly reduced. Moreover, a noisy prioritized experience replay (PER) algorithm is proposed to improve the exploring rate of control policy. The effectiveness of applying the proposed deep learning-based control is validated by executing a number of simulation and experimental case studies. The simulation result shows that the proposed DRL method outperforms the vanilla PER algorithm in terms of training speed. Experimental videos are also uploaded, and the corresponding results confirm that the proposed strategy is able to fulfill the autonomous exploration mission with improved motion planning performance, enhanced collision avoidance ability, and less training time.

Tracking Control of Self-Restructuring Systems: A Low-Complexity Neuroadaptive PID Approach With Guaranteed Performance

This article investigates the tracking control problem for a class of self-restructuring systems. Different from existing studies on systems with fixed structure, this work focuses on systems with varying structures, arising from, for instance, biological self-developing, unconsciously switching, or unexpected subsystem failure. As the resultant dynamic model is complicated and uncertain, any model-based control is too costly and seldom practical. Here, we explore a nonmodel-based low-complexity proportional-integral-derivative (PID) control. Unlike traditional PID with fixed gains, the proposed one is embedded with neural-network (NN)-based self-tuning adaptive gains, where the tuning strategy is analytically built upon system stability and performance specifications, such that transient behavior and steady-state performance are ensured. Both square and nonsquare systems are addressed by using the matrix decomposition technique. The benefits and feasibility of the proposed control method are also validated and confirmed by the simulations.

Adaptive Multigradient Recursive Reinforcement Learning Event-Triggered Tracking Control for Multiagent Systems

This article proposes a fault-tolerant adaptive multigradient recursive reinforcement learning (RL) event-triggered tracking control scheme for strict-feedback discrete-time multiagent systems. The multigradient recursive RL algorithm is used to avoid the local optimal problem that may exist in the gradient descent scheme. Different from the existing event-triggered control results, a new lemma about the relative threshold event-triggered control strategy is proposed to handle the compensation error, which can improve the utilization of communication resources and weaken the negative impact on tracking accuracy and closed-loop system stability. To overcome the difficulty caused by sensor fault, a distributed control method is introduced by adopting the adaptive compensation technique, which can effectively decrease the number of online estimation parameters. Furthermore, by using the multigradient recursive RL algorithm with less learning parameters, the online estimation time can be effectively reduced. The stability of closed-loop multiagent systems is proved by using the Lyapunov stability theorem, and it is verified that all signals are semiglobally uniformly ultimately bounded. Finally, two simulation examples are given to show the availability of the presented control scheme.

Finite-Time Robust Intelligent Control of Strict-Feedback Nonlinear Systems With Flight Dynamics Application

The tracking control is investigated for a class of uncertain strict-feedback systems with robust design and learning systems. Using the switching mechanism, the states will be driven back by the robust design when they run out of the region of adaptive control. The adaptive design is working to achieve precise adaptation and higher tracking precision in the neural working domain, while the finite-time robust design is developed to make the system stable outside. To achieve good tracking performance, the novel prediction error-based adaptive law is constructed by considering the estimation performance. Furthermore, the output constraint is achieved by imbedding the barrier Lyapunov function-based design. The finite-time convergence and the uniformly ultimate boundedness of the system signal can be guaranteed. Simulation studies show that the proposed approach presents robustness and adaptation to system uncertainty.

Event-Triggered Adaptive Control of Uncertain Nonlinear Systems With Composite Condition

This article concentrates on the event-based collaborative design for strict-feedback systems with uncertain nonlinearities. The controller is designed based on neural network (NN) weights adaptive law. The controller and NN weights adaptive law are only updated at the triggering instants determined by a novel composite triggering threshold. Considering the conservativeness of event condition, the state-model error is integrated into constructing the composite condition and NN weights adaptive law. In the context of the proposed mechanism, the requirements of system information and the allowable range of event-triggering error are relaxed. The number of triggering instants is greatly reduced without deteriorating the system performance. Moreover, the stability of the closed-loop is proved by the Lyapunov method following time-interval and sampling instants. Simulation results show the effectiveness of the scheme proposed in this article.

Octopus-Inspired Microgripper for Deformation-Controlled Biological Sample Manipulation

Predators in nature grip their prey in different ways, which give innovational ideas of gripping approaches in industrial applications. Octopus performs flexible gripping with the help of vacuum grippers, suction cups, which inspired a new type of microgripper for biological sample micromanipulation. The proposed gripper consists of a glass pipette and a pump driven by a step-motor. The step-motor is controlled with adaptive robust control to adjust the gripping pressure applied on the biological sample. A dynamic model is developed for the biological sample aiming for better deformation control performance. A visual detection algorithm is developed for data processing to identify the parameters in the dynamic model and the detection result of visual algorithm is also used as feedback of adaptive robust control, which diminishes the negative influence of parameter and model uncertainties. Zebrafish larva was used as the testing sample for experiment and the corresponding parameters were identified experimentally. The experimental results correlated well with the model predicted deformation curve and visual detection algorithm provided promising accuracy, which is less than [Formula: see text]. Adaptive robust control provides fast and accuracy response in point-to-point deformation testing, and the average responding time is less than 30 s and the average error is no larger than 1 pixel.

Output Feedback Control of Micromechanical Gyroscopes Using Neural Networks and Disturbance Observer

This article addresses the output feedback control of micromechanical (MEMS) gyroscopes using neural networks (NNs) and disturbance observer (DOB). For the unmeasured system states, the state observer and the high gain observer are constructed. The adaptive NNs are investigated to approximate the nonlinear dynamics, including the known nominal terms and the system uncertainties caused by environmental fluctuations. For the time-varying disturbances, the DOB is utilized. The sliding mode control is employed to enhance the robustness. Through simulation verification, the output feedback control using NNs and DOB can adapt to the dynamics of MEMS gyroscope with unmeasured system speed, while an expected effective tracking performance is obtained in the presence of unknown system nonlinearities and external disturbances.

Neural-Network-Based Sliding-Mode Control of an Uncertain Robot Using Dynamic Model Approximated Switching Gain

In this article, a new neural-network-based sliding-mode control (SMC) of an uncertain robot is presented. The distinguishing characteristic of the proposed control scheme is that the switching gain is designed as a dynamic model approximated value, which is handled by using the neural-network strategy to adapt the unknown dynamics and disturbances. In the presented control scheme, the modeling information of the robotic system is not required and only one parameter is required to be estimated in each joint of the robotic system. Subsequently, the Lyapunov method is utilized to prove that the trajectory tracking errors will eventually converge to a neighborhood of zero. Finally, the contrast simulation studies reveal that with the proposed control scheme, the problems of chattering and high-speed switching of control input, which takes place in a conventional SMC, can be addressed, and a satisfactory control precision is guaranteed.