![]() ![]() Naturally, the implementation costs of real platoons increase as the number of vehicles in the platoon increase. The latter implies the setting up of a communication network to send and receive information among vehicles. To achieve a string stable configuration with linear control, the agents must posses, at least, either a measurement of their speed or be able to receive the position and/or speed of the platoon leader. Indeed, to achieve the platooning goal, the vehicles must be appropriately equipped with distance sensors, a computing unit that will determine the acceleration or braking actions and a power stage to manipulate the movement. As the study and design of control schemes is of high theoretical complexity, their practical implementation aspects are no less so, and must be considered. The theoretical aspect of such problem has been widely studied, including configurations where string stability is guaranteed. These types of configurations are suitable for modeling AHS where the vehicles will rarely overtake other members of the platoon, unless they are merging or departing from the formation. In particular we are interested in a platoon of vehicles that autonomously navigates a track in one direction, maintaining a predetermined inter-vehicle distance while traveling at a desired speed. Preliminary results in this direction where previously repoted in. In this work we are interested in developing an experimental platform for MAS problems where the agents are vehicles that can move in one dimension. Such undesirable phenomena is known as string instability. In such a scenario, platooning schemes able to achieve a tight formation with a constant speed reference, may suffer from the amplification of disturbances or local errors, as they propagate along the chain of vehicles, inducing performance and safety issues that are intensified as the number of vehicles of the platoon increases. Scalability is a key aspect in applications such as automated highway systems (AHS), in which a platoon of vehicles are automated to navigate in a highway. A given control system that exhibits an acceptable behavior for a MAS composed of a few agents could cause poor performance for a MAS with a larger number of agents. In this context, scalability refers to the behavior of the MAS when the number of agents increases. This motivates the development of experimental setups suitable to study such systems and their most relevant aspects, where scalability issues rank as one of the most important. Given the nature of these type of systems, whose complexity increases with the number of members and interactions, the study of their coordinated behavior is a challenging task, as is the design of appropriate control systems and their implementation. Industrial processes, transportation systems, and energy systems, to name a few examples, are among the main areas of impact that can be potentially optimized through the design of control algorithms for systems comprised of several dynamical agents, aiming to work in a coordinated fashion. ![]() In recent years, advances on control systems and wireless communications technologies have facilitated the implementation of complex multi-agent systems (MAS) applications. Additionally, we illustrate with experimental results some of the teaching activities that the platform is capable of performing. The agents are built from low cost components and programmed with open source software, enabling teaching experiences and experimental work with a larger number of agents that would otherwise be possible with other existing solutions. The individual agents can be used for simple PID experiments in a classroom or laboratory setting, while a collection of agents are capable of performing decentralized platooning with cooperative adaptive cruise control in a variety of settings, the latter being the main goal of the platform. The platform consists of train-like autonomous agents equipped with local speed estimation, distance sensing to their nearest predecessor, and wireless communications with other agents and a central coordinator. In this paper, we present the development of a low-cost multi-agent system experimental platform for teaching, and research purposes. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |