Search

In this article, a cooperative multitechnology simultaneous localization and signal mapping (CM-SLASM) technique is proposed to improve the signal map accuracy and to build on-the-fly a signal map to be used also in indoor environments where conditions can change over time. Moreover, the CM-SLASM combines wireless fidelity (WiFi), ultra wideband (UWB), and light detection and ranging (LIDAR) signals to improve positioning estimation by sharing information and cooperation among vehicles through vehicle-to-vehicle (V2V) communication links. In particular, LIDAR-based distance between vehicles is shared among neighbor vehicles to improve the vehicle positioning estimated by an extended Kalman filter (EKF) where WiFi fingerprinting is combined with UWB multilateration. The overall solution where EKF estimation allows to building of more precise signal MAP is validated by simulation in a defined indoor scenario where vehicles equipped with different percentages of LIDAR, and different quantities of UWB and WiFi emitters have been considered. The proposed strategy has been validated through an extensive simulation campaign in various scenarios of interest and through a real-world experiment conducted in a laboratory test environment.

In this paper, platoons of autonomous vehicles operating in urban road networks are considered. From a methodological point of view, the problem of interest consists of formally characterizing vehicle state trajectory tubes by means of routing decisions complying with traffic congestion criteria. To this end, a novel distributed control architecture is conceived by taking advantage of two methodologies: deep reinforcement learning and model predictive control. On one hand, the routing decisions are obtained by using a distributed reinforcement learning algorithm that exploits available traffic data at each road junction. On the other hand, a bank of model predictive controllers is in charge of computing the more adequate control action for each involved vehicle. Such tasks are here combined into a single framework: the deep reinforcement learning output (action) is translated into a set-point to be tracked by the model predictive controller; conversely, the current vehicle position, resulting from the application of the control move, is exploited by the deep reinforcement learning unit for improving its reliability. The main novelty of the proposed solution lies in its hybrid nature: on one hand it fully exploits deep reinforcement learning capabilities for decision-making purposes; on the other hand, time-varying hard constraints are always satisfied during the dynamical platoon evolution imposed by the computed routing decisions. To efficiently evaluate the performance of the proposed control architecture, a co-design procedure, involving the SUMO and MATLAB platforms, is implemented so that complex operating environments can be used, and the information coming from road maps (links, junctions, obstacles, semaphores, etc.) and vehicle state trajectories can be shared and exchanged. Finally by considering as operating scenario a real entire city block and a platoon of eleven vehicles described by double-integrator models, several simulations have been performed with the aim to put in light the main features of the proposed approach. Moreover, it is important to underline that in different operating scenarios the proposed reinforcement learning scheme is capable of significantly reducing traffic congestion phenomena when compared with well-reputed competitors.

This article proposes a decentralized controller for a swarm of agents that generates hypotrochoidal and epitrochoidal curves in a 3-D space. The dynamics of the agents is modeled as a double integrator where the control input, i.e., the acceleration profile, contains two terms that account for the interactions between the agents in terms of position and velocity, respectively. The term related to the interaction between velocities is weighted by a skew-symmetric matrix responsible for coupling the velocities displacements. Such a matrix defines an axis of rotation centered on the centroid of the swarm, around which the swarm organizes itself to generate the desired trajectories, which can be selected by properly setting two free parameters of the model. In particular, the motion of each agent is analyzed by separating its dynamics along the rotation axis from that in the perpendicular plane. It is shown that the agents achieve consensus or periodic motion along the axis of rotation while performing the desired evolutions in the perpendicular plane. Numerical simulations are presented to illustrate the proposed framework.