This paper deals with the decentralized predefined-time leaderless consensus formation control problem for nonholo-nomic multi -agent systems. First, the predefined-time average consensus problem is investigated. Distributed observers, which interact with each other via an undirected communication topology, are designed to estimate, in a predefined-time, the average of their initial state using only local information. Then, based on these estimates, a predefined-time Vector-Field-Orientation (VFO) controller is applied, which guarantees that the multi-agent system achieves a desired formation. The VFO controller enables one to predict the paths for each agent and to avoid static obstacles. Some numerical results illustrate the interest of the proposed predefined-time leaderless consensus formation controller.
This paper deals with the trajectory tracking problem for autonomous mobile robots under discrete-time measurements, assuming that the available measurements are the position and orientation of the mobile robot. With the mentioned measurements, an implicit discretization of the robust exact filtering differentiator is proposed to estimate the position and velocity of the mobile robot despite the presence of measurement noise while considering the sampling effect. Consequently, an output feedback controller is derived using the estimates from the discrete-time filtering differentiator, guaranteeing the closed-loop system's stability despite noise and discrete measurements. Numerical results illustrate the effectiveness of the proposed scheme.
This paper studies the cooperative navigation problem for multirobot systems with guaranteed collision avoidance and connectivity maintenance, under disturbances and input saturation. To achieve robust cooperative navigation while ensuring safety requirements, robust control barrier function (RCBF) is utilized, with the helps from disturbance observer for flexibility and asymptotically state consensus. The safety condition in QP problem is given in a unique and distributed fashion. A switching topology is constructed to avoid unnecessary connectivity constraints. Simulation results help verify the effective of our proposed strategy.
This paper merges for the first time two recent forms of adaptive control techniques, namely two-layer model reference adaptive control (MRAC) and funnel control, also known as tube-based control. Two-layer MRAC allows the user to set the rate of convergence of the tracking error arbitrarily high without altering the user-defined reference model or relying on barrier functions. Funnel control is a technique whereby the tracking error is constrained within bounds that can be progressively tighter. The proposed approach merges the benefits of both techniques and, for the first time in the literature on funnel control, explicitly regulates the diameter of the funnel through some user-defined adaptive law. The proposed results are tested by simulating payload delivery missions using multi-rotor unmanned aerial vehicles (UAVs) in the high-fidelity environment PyChrono.
Self-collision checking plays an important role in robot motion planning. In sampling-based motion planning methods, collision checking is performed multiple times, so this procedure should be accurate and fast. Collision checking is often addressed in robotics through the application of machine learning methods or a relatively straightforward multilayer perceptron within a low-dimensional feature space. In this paper, we focus on incorporating positional encoding, commonly utilized in computer graphics, into the input vector used in the binary classification task. We investigate the enhancement in classification accuracy by utilizing this technique in self-collision checking. Our findings indicate that positional encoding contributes to improved learning of high-frequency functions and provides a more accurate representation of higher-frequency details within the trained relation.
