Carlos Sagüés

Eduardo Montijano, Juan I. Montijano, Carlos Sagues

In this paper we analyze the use of Chebyshev polynomials in distributed consensus applications. We study the properties of these polynomials to propose a distributed algorithm that reaches the consensus in a fast way. The algorithm is expressed in the form of a linear iteration and, at each step, the agents only require to transmit their current state to their neighbors. The difference with respect to previous approaches is that the update rule used by the network is based on the second order difference equation that describes the Chebyshev polynomials of first kind. As a consequence, we show that our algorithm achieves the consensus using far less iterations than other approaches. We characterize the main properties of the algorithm for both, fixed and switching communication topologies. The main contribution of the paper is the study of the properties of the Chebyshev polynomials in distributed consensus applications, proposing an algorithm that increases the convergence rate with respect to existing approaches. Theoretical results, as well as experiments with synthetic data, show the benefits using our algorithm.

Eduardo Sebastian, Thai Duong, Nikolay Atanasov et al.

The graph identification problem consists of discovering the interactions among nodes in a network given their state/feature trajectories. This problem is challenging because the behavior of a node is coupled to all the other nodes by the unknown interaction model. Besides, high-dimensional and nonlinear state trajectories make it difficult to identify if two nodes are connected. Current solutions rely on prior knowledge of the graph topology and the dynamic behavior of the nodes, and hence, have poor generalization to other network configurations. To address these issues, we propose a novel learning-based approach that combines (i) a strongly convex program that efficiently uncovers graph topologies with global convergence guarantees and (ii) a self-attention encoder that learns to embed the original state trajectories into a feature space and predicts appropriate regularizers for the optimization program. In contrast to other works, our approach can identify the graph topology of unseen networks with new configurations in terms of number of nodes, connectivity or state trajectories. We demonstrate the effectiveness of our approach in identifying graphs in multi-robot formation and flocking tasks.

Irene Perez-Salesa, Rodrigo Aldana-Lopez, Carlos Sagues

Visual object tracking under challenging conditions of motion and light can be hindered by the capabilities of conventional cameras, prone to producing images with motion blur. Event cameras are novel sensors suited to robustly perform vision tasks under these conditions. However, due to the nature of their output, applying them to object detection and tracking is non-trivial. In this work, we propose a framework to take advantage of both event cameras and off-the-shelf deep learning for object tracking. We show that reconstructing event data into intensity frames improves the tracking performance in conditions under which conventional cameras fail to provide acceptable results.

Rodrigo Aldana-López, Rosario Aragüés, Carlos Sagüés

Target tracking is a popular problem with many potential applications. There has been a lot of effort on improving the quality of the detection of targets using cameras through different techniques. In general, with higher computational effort applied, i.e., a longer perception-latency, a better detection accuracy is obtained. However, it is not always useful to apply the longest perception-latency allowed, particularly when the environment doesn't require to and when the computational resources are shared between other tasks. In this work, we propose a new Perception-LATency aware Estimator (PLATE), which uses different perception configurations in different moments of time in order to optimize a certain performance measure. This measure takes into account a perception-latency and accuracy trade-off aiming for a good compromise between quality and resource usage. Compared to other heuristic frame-skipping techniques, PLATE comes with a formal complexity and optimality analysis. The advantages of PLATE are verified by several experiments including an evaluation over a standard benchmark with real data and using state of the art deep learning object detection methods for the perception stage.

Javier Tardós, Rosario Aragues, Carlos Sagüés et al.

Multi robot teams composed by ground and aerial vehicles have gained attention during the last years. We present a scenario where both types of robots must monitor the same area from different view points. In this paper we propose two Lloyd-based tracking strategies to allow the ground robots (agents) follow the aerial ones (targets), keeping the connectivity between the agents. The first strategy establishes density functions on the environment so that the targets acquire more importance than other zones, while the second one iteratively modifies the virtual limits of the working area depending on the positions of the targets. We consider the connectivity maintenance due to the fact that coverage tasks tend to spread the agents as much as possible, which is addressed by restricting their motions so that they keep the links of a Minimum Spanning Tree of the communication graph. We provide a thorough parametric study of the performance of the proposed strategies under several simulated scenarios. In addition, the methods are implemented and tested using realistic robotic simulation environments and real experiments.