International Journal of Robust and Nonlinear Control最新文献

This article studies the bipartite tracking control problem for multi-agent systems with a distributed event-triggered estimator under prescribed performance. Firstly, a distributed estimator is developed, allowing agents not directly connected to the leader direct access to the desired consistency change trajectory. Meanwhile, a dynamic event-triggered mechanism is introduced to eliminate the need for continuous monitoring of neighboring agents' information. Then, for the controller-to-actuator channel, an event-triggered mechanism based on the error information is designed to save communication resources. Furthermore, a prescribed performance control method with a variety of adjustable preset and non-monotonic boundaries is introduced into the design of the fuzzy adaptive controller, which avoids the process of redesigning the control after adjusting the boundaries. It is demonstrated that all signals in the closed-loop systems are semiglobally uniformly ultimately bounded, and Zeno behavior is excluded in the event-triggered process. Finally, simulation results validate the effectiveness of the proposed scheme.

In the natural world, coordinated behaviors like group migration and collective defense exemplify the capacity of biological populations to process spatial information and adapt their actions correspondingly. Drawing inspiration from these natural phenomena, researchers have introduced diverse multi-agent cooperative control strategies, such as fixed neighborhood algorithms, fixed number algorithms, and farthest neighborhood algorithms. Nevertheless, these approaches encounter obstacles including sluggish convergence rates, limited resilience to interference, and constrained interpretability. To overcome these limitations, this paper introduces an innovative distributed multi-agent cooperative control algorithm leveraging an attention mechanism. The algorithm dynamically modulates the information perception weights of agents, facilitating effective coordination in intricate environments. Furthermore, it integrates a pioneering obstacle avoidance strategy, guaranteeing resilient performance in dynamic and densely populated environments. In comparison to existing approaches, the proposed algorithm demonstrates accelerated convergence, enhanced interference resistance, and improved scalability in formation control, all while preserving theoretical interpretability. This study offers a fresh perspective and innovative methodology for advancing cooperative control in multi-agent systems, with wide-ranging applications in fields like robotics and autonomous systems.

This paper addresses the optimal observer design problem of continuous-time Markovian jump systems with sampled signals. A new observer deeply depending on mode separations is developed to deal with its sampled switching signals, which overcomes the difficulties of sampled and nonsampled signals contained simultaneously. Compared with the traditional methods, not only the wider application scope but also less conservatism can be ensured. Moreover, a novel optimization problem about mode separations is proposed to give the best mode separation, whose locally optimal solution can be obtained by employing the DDPG algorithm. Compared with the existing method without any optimization, the observer performance can be further improved in terms of lower energy consumption. A numerical simulation is conducted to verify the utility and superiority of the proposed methods.

We investigate adaptive optimal and non-optimal multi-lane fusion control for a two-dimensional (2-D) plane vehicle platoon under angle constraints and prescribed performance. Most existing optimal control methods are restricted to optimization criteria and lack flexibility in adjusting control structures. In contrast, the proposed strategy enables both optimal and non-optimal control by employing a layered controller design. One part uses sliding mode control to achieve the desired prescribed performance while adhering to angle constraints. The other part incorporates optimal control, utilizing an adaptive dynamic programming (ADP) method with the actor-critic (AC) strategy to minimize the performance index function related to sliding-mode surfaces and control signals. The sliding mode component alone achieves non-optimal control, ensuring multi-lane fusion and vehicle platoon stability. When combined with the optimal control component, the vehicle platoon minimizes the performance index function, thus achieving optimal multi-lane fusion control. Besides, the unmodeled dynamics in the follower vehicle model are approximated by the radial basis function neural networks (RBFNNs) with the updating laws derived from the gradient descent method. Finally, the effectiveness of the proposed strategy is verified through simulations and comparisons.

This article investigates annular finite-time stability (AFTS) and stabilization in probability for nonlinear stochastic Markov jump systems with impulsive effects. First, a new notion called AFTS in probability is introduced, which is weaker than AFTS in mean-square sense, and a new proposition that the AFTS in mean square sense implies the AFTS in probability under general assumptions is given by constructing inverse function and Burkholder-Davis-Gundy inequality. Second, by the inverse technique of Lyapunov function and dividing the system state at distinct impulsive instants, the testing criterion for AFTS in probability is derived. Moreover, sufficient conditions for state feedback AFT-stabilizing controller design are obtained, and a set of possible control laws is given in detail. Finally, a numerical example is provided to verify the validity of our proposed results.

Land–air hybrid locomotion robots are popular because of their exceptional cross-domain operational capabilities, although control challenges during air-to-ground transitions remain a key research focus. Flying quadrupedal robots exhibit excellent terrain adaptability, rapid mobility, and remarkable static load performance during flight, making them highly valuable for research and development. To increase adaptability to complex terrains, we model the robot's legs as a mass-spring-damper system and propose a compliant adaptation strategy based on plantar pressure information for irregular surfaces. Landing is divided into five distinct stages, enabling a smooth transition from flight to ground locomotion. To increase system robustness during landing, we introduce a reinforcement learning-based self-tuning generalized PID control framework. In this framework, the critic neural network (NN) approximates the cost function, and the actor NN dynamically adjusts the generalized PID gains via feedback from the critic. The optimized control inputs drive the system towards optimal performance. To address actuator input saturation, we reformulate the problem with a smooth hyperbolic tangent function, transforming it into a variable-gain control problem with a known control direction. The Nussbaum function mitigates the adverse effects of input saturation. This approach systematically adjusts PID gains for flying quadrupedal robots, yielding a simple yet computationally efficient framework that avoids extensive parameter tuning. Finally, the stability of the closed-loop system is proven via Lyapunov theory. The numerical simulations conducted on a flying quadrupedal robot prototype—integrating a one-dimensional vectoring thrust VTOL platform with a Mini Cheetah quadruped—further validate the effectiveness of the proposed strategy.

In this paper, the data-driven control issue for a class of nonlinear networked control systems (NCSs) with time-varying delays is investigated. To achieve the tracking control of the nonlinear NCS, an event-triggered model-free adaptive predictive control (MFAPC) strategy is developed. First, an equivalent partial-form dynamic linearization data model is established based on the time-varying pseudo gradient, which is calculated by adopting the input/output (I/O) data of the nonlinear NCS. Next, the networked predictive control strategy is applied to deal with the negative effects of time-varying delays on system performance in the sensor-to-controller (S–C) and controller-to-actuator (C–A) channels. Subsequently, two dynamic event-triggered control mechanisms are adopted to balance the expected system performance and consumption of network resources. Moreover, the stability criterion of the closed-loop nonlinear NCS is provided, and zero tracking error can be proved. In the end, the simulation example is performed to demonstrate the validity and superiority of the developed strategy.

For the switching-mode systems, the stability is closely related to the symbolic sequence of system states. Utilizing a symbolic sequence transformation technique, this paper designs a third-order twisting algorithm (TA) and studies its finite-time stability. A stability condition of second-order TA is established, and an analytic expression of convergence time for second-order TA is deduced. By analyzing the importance mechanism of Platform Phase in the symbolic sequence, a stabilized third-order TA is developed, and a new strategy on parameter selection is given to achieve the finite-time stability of third-order TA. The derived convergence times starting from different initial states reveal the quantitative influence of control parameters on dynamic performances of TA, providing guidance for the design of settling time-customized TA. The application of TA in DC–DC converters is presented, and the comparative experiments demonstrate the superior performances of the proposed method in terms of response time, control accuracy, and disturbances rejection.

This paper aims to solve the optimal consensus problem for homogeneous systems of general linear dynamics and a cost function. Specifically, we present a new method rooted in LQ optimal control theory and observer design. This method is fundamentally different from the existing consensus control algorithms, which rely on nonzero eigenvalues of the communication topology to determine the distributed controller. The analytical solution for the distributed controller is derived via Riccati equations and observers, which is parallel to the classical optimal control theory. Theoretical analysis and a simulation example demonstrate that the proposed optimal consensus method achieves significantly faster convergence than conventional approaches. Notably, this framework can be extended to tackle consensus problems in heterogeneous multi-agent systems.