This article addresses the problem of prescribed performance tracking control for uncertain differential-drive mobile robots equipped with current sensorless electric motors. An improved line-of-sight distance control method with a desired heading angle switching strategy is proposed, which eliminates the singularity problem encountered in previous works when the line-of-sight distance approaches zero. In addition, an open issue of mobile robots control under unavailable motor current (or torque) and completely unknown motor parameters is tackled by combining differential homeomorphism transformations with a low-complexity prescribed performance control approach. Furthermore, the limitations imposed by initial high-order state errors in low-complexity control methods are overcome by introducing a time and space dependent amplifier. The proposed method guarantees the convergence of the position tracking error to the desired precision within the prescribed time, with rigorous proof provided. The effectiveness of the method is demonstrated through simulations and experiment comparisons.

The existing shared control schemes and automatic control schemes are almost designed separately. Based on the designed constrained function and shared-constrained function, this paper proposes a universal control framework for high-order strict-feedback systems that can achieve safe tracking control with very low algorithm complexity regardless of whether the reference signal exceeds safety constraints, combining the flexibility of manual operating with the persistence and security of automatic control. A universal safe tracking control theorem is presented to prove the security of system operation. Furthermore, based on the proposed universal safe tracking control method, a novel time-independent path following control method with lane constraints is proposed for nonholonomic mobile robots by converting lane constraints into heading angle constraints, thus achieving automatic/auxiliary adaptive cruise control and lane keeping. Simulation and experimental results are provided to demonstrate the practical applicability of this method. Note to Practitioners—This work addresses the issue of the reference signal exceeding the safety constraint due to planner errors (in autonomous control) or human operator mistakes (in shared control). Compared to existing control methods, the proposed architecture is not only applicable to safe tracking control in both autonomous and shared control scenarios, but also features extremely low algorithmic complexity, making it highly suitable for practical deployment. Unlike traditional shared control systems, where human operators directly manipulate system inputs, the proposed method allows operators to control the system by adjusting the reference signal. This ensures that feedback control is always present, thereby reducing the difficulty of manipulation. Moreover, in contrast to common trajectory tracking methods, the proposed time-independent path following approach holds greater practical value, ultimately achieving adaptive cruise control and lane-keeping for mobile robots. To ensure repeatability, our codes are open sourced on github:https://github.com/Mudianrui/USC-and-app-in-NMRs.git. Experimental videos can be accessed throughhttps://youtu.be/VrsISKkJOCA

This paper studies the Lyapunov-based adaptive neural network optimized tracking control problem for a class of unmanned fire fighting robots. Firstly, by reasonably simplifying the unmanned fire fighting robot (UFFR) and combining it with its actual working scene, a novel system model is created that takes into consideration both system uncertainties and external disturbances, including unknown friction factors and drag force. Then, the optimized tracking control scheme for the UFFR is devised by integrating both adaptive neural networks and the backstepping technique. The objective of introducing adaptive neural network technique is to overcome the challenge posed by solving the Hamilton-Jacobi-Bellman (HJB) equation. Based on Lyapunov stability theory, it is demonstrated that all signals in the closed-loop system are semi-globally ultimately bounded and the output variables follow the reference signals to the desired accuracy. In the end, to validate the effectiveness of our designed control scheme, numerical simulations and practical platform experiments have been conducted. To ensure repeatability, our codes are open sourced on Github: https://github.com/JiannanChen/RL-based-OBC-of-UFFR.git

This paper investigates the lateral and longitudinal platoon control problem under user-specified lane and inter-vehicle spacing constraints. By modeling in the Frenét Frame, the lateral and longitudinal movements of the vehicles are decomposed. A novel lateral control strategy is proposed to strictly enforce a preset lane departure accuracy by transforming lane constraints into heading angle constraints. In the longitudinal control, considering the presence of a non-ideal leading vehicle, a prescribed performance controller is designed to regulate its velocity. Additionally, a longitudinal control strategy based on a double-ended smooth transition function is proposed to mitigate the effects of non-zero initial errors and restore the standard constant time headway policy after a preset time. To guarantee practical fixed-time stability, a continuous variable exponent coefficient fixed-time sliding surface is constructed, and adaptive sliding mode controllers are designed. The effectiveness of the proposed method is validated through both simulations and experiments. The source code is available on GitHub (https://github.com/Mudianrui/FSC-VPuLLC.git) to support further research on lateral-longitudinal platoon control.

This article focuses on the fully distributed dual-event triggered leader-following consensus problem of heterogeneous multiagent systems (MASs) with unknown leader input and actuator fault. A hierarchical triggered control framework is developed for the MASs. First, in the upper network layer, event-triggered observers are designed to reconstruct the leader’s information by using the states exchanged intermittently among the neighboring observers. Then, in the lower physical layer, an adaptive fault-tolerant controller is designed, whose update instant can be directly computed based on the received upper layer information. This not only has the potential to further reduce the update frequency of controller, but also avoid the continuous monitoring on measurement errors. Besides, utilizing this control framework can prevent the faults from being propagated along the communication network. Finally, a numerical example is provided to verify the effectiveness of theoretical results.

To address the classification problem of short time-window steady-state visual evoked potentials (SSVEP), a novel deep-convolutional neural network (CNN) fused with residual squeeze and excitation blocks (RSEs) and multiscale convolutions is proposed. Given the difficulty in distinguishing frequency domain features of short time-window signals, AttentCNN-Multiscale begins with a filter bank (FB)-based time-domain feature extraction module. The FB comprises several sixth-order Butterworth filters with varying bandpass ranges. Then the feature tensors extracted by these filters are aggregated using a CNN with RSEs. For further feature learning, four 2-D CNNs and a multiscale convolution module are employed, with the final output generated through an adaptive fully connected layer. To demonstrate the effectiveness and superiority of AttentCNN-Multiscale, extensive experiments and comparisons are conducted on two large public datasets and our dataset. Additionally, a novel bionic intelligent soft gripper is designed and integrated with the proposed AttentCNN-Multiscale network to form a closed-loop system, enabling different grasping functionalities for various objects and demonstrating the application potential of the network in medical rehabilitation. To ensure reproducibility, the source code for AttentCNN-Multiscale is available on Github: https://github.com/raow923/AttentCNN-Multiscale.

This paper considers the tracking control problem for unmanned fire fighting robots subject to both full-state constraints and input saturation. First, a system model is developed that incorporates system internal uncertainties, external disturbances, input saturation, and actuator faults. Then, to address the full-state constraint problem, the original constrained system is transformed into an equivalent unconstrained one by using a new state-dependent transformation function. In addition, to solve asymmetric time-varying constraints on the control input, another new transformation function is also designed. In the end, based on the transformed system, a novel adaptive control scheme is proposed utilizing the backstepping recursive method and first-order filters. It is demonstrated that all signals in the closed-loop system are semi-globally ultimately bounded, and the output variables accurately track the reference signals while satisfying both full-state constraints and input saturation. To validate the effectiveness of our designed control scheme, numerical simulations have been conducted. To ensure repeatability, our codes are open sourced on github: https://github.com/JiannanChen/ATControlUFFR-FullStateConstraintInputSaturation.git.

This article studies the modeling and tracking control problems for a class of towed unmanned fire fighting robots. Considering that no similar modeling results exist, we take the lead in building a novel system model that takes into consideration both system uncertainties and external disturbances, including unknown friction factors and drag force. Then, to compensate for the adverse effects of system uncertainties and external disturbances, a novel robust control algorithm is proposed, which utilizes adaptive control and scaling techniques. Moreover, innovative predefined performance functions are designed to ensure that tracking processes meet predefined transient and steady-state requirements. Unlike most of the existing works, our predefined time performance function has the advantage that the convergence time and convergence accuracy can be arbitrarily changed. In the end, a novel robust adaptive control scheme with predefined time and precision tracking is designed using the backstepping recursive method. Based on Lyapunov stability theory, it is demonstrated that all signals in the closed-loop system are ultimately bounded, and both predefined transient and steady-state processes are never violated. To validate the effectiveness of this proposed control scheme, numerical simulations and practical platform experiments have been conducted. To ensure repeatability, our codes are open sourced on Github: https://github.com/JiannanChen/Modeling-and-Robust-Control-UFFR.git.