We investigate the consensus problems for uncertain multi-agent systems (MASs) via a neural network (NN) adaptive control approach. However, it is quite difficult because of the existence of NN approximation errors. Unlike existing works that use discontinuous functions to eliminate the effects of NN approximation errors, a novel continuous function is proposed by utilizing the idea of time varying boundary layer technique. Firstly, we design a continuous NN adaptive controller with dynamic coupling strengths for the case with time invariant topology. It is shown that asymptotical consensus can be achieved without requiring any global information. Secondly, we consider time varying communication networks and design a continuous NN adaptive controller with fixed coupling strength. It is shown that the consensus error is ultimately uniformly bounded if the average dwell time is larger than a positive scalar. Finally, two examples are studied to verify the theoretical results, respectively.

An improved path planning algorithm based on dynamic window approach (DWA) is proposed to optimize real-time paths for agents moving in complex environments with plenty of static and/or dynamic obstacles, path constraints, and performance constraints. In order to improve the accuracy of the evaluation function and the adaptability of the agents to different obstacle environments, the weight of the evaluation function is ameliorated by designing adaptive laws in this article. Meanwhile, aiming at the defect of local minima in DWA algorithm, a low-speed steering strategy is presented to help the robot escape the U-shaped trap. Especially, in the case of dynamic obstacle environment, a predictive scheme is carried out for the agents to reach the destination safely by fitting the trajectory of the dynamic obstacles to calculate their possible future positions, and then setting these positions as virtual obstacles. By comparison, some simulations and experiments are illustrated to show that the proposed algorithms are superior to the DWA algorithm in terms of running time, path length, and security to a large extent.

Space-based Earth observation is now developing rapidly because of the advantages in coverage, convenience, and flexibility. However, it is difficult for one single satellite to realize the quick observation of wide areas. In order to catch all the significant information of a wide area in a short time, multi-satellite observation mission would be proposed. In this article, a mission planning method for online multi-satellite wide area observation is established to serve future multi-satellite observation missions. Firstly, a method for area division is proposed, and the whole area is divided into subareas. Then the multi-satellite observation path planning is realized by a strategy of path deduction. After that, a remaining time allocation method to maximize the observation gazing time of each subarea is proposed. Finally, the algorithms for the whole mission planning process are provided. Numerical simulations show that the mission planning method is able to ensure the complete coverage of different wide target areas, with high reliability and low computational complexity.

In this article, relative output information is applied to tackle the consensus control problem of linear multiagent systems (MASs) under biasing faults and uncertainty in the parameters. An auxiliary system is first introduced to reconstruct the fault signal, and a class of observers is proposed to simultaneously estimate the current states and faults of the augmented MASs. According to the estimated information, distributed ${\mathcal{H}}_{\infty}$ and ${\mathcal{H}}_{2}$ consensus controllers are proposed, respectively, within this design framework. By utilizing the consensus region-based analysis technique, some sufficient criteria are derived to achieve ${\mathcal{H}}_{\infty}$ and ${\mathcal{H}}_{2}$ consensus, respectively, when a piece of spectrum information of the Laplacian matrix related to communication topology is involved. The Laplacian matrix’s spectrum can be fully decoupled from the observer and controller design process in the case of only actuator faults. Compared with the existing methods, the proposed algorithms almost exclusively depend on the relative output information with little global information of the graph, thereby making the solution more scalable and applicable to practical situations. Numerical simulations ultimately demonstrate the validity of the proposed results.