Decentralized Receding Horizon Control for Cooperative Multiple Vehicles Subject to Communication Delay

摘要

IN this paper, a new approach is proposed for the decentralizednreceding horizon control (DRHC) of multiple cooperativenvehicles with the possibility of communication failures leading tonlarge intervehicle communication delay. Such large communicationndelays can lead to poor performance and even instability. Thenneighboring vehicles exchange their predicted trajectories at eachnsample time to maintain the cooperation objectives. It is assumed thatnthe communication failure is partial in nature, which in turn leads tonlarge communication delays of the exchanged trajectories. Thenproposed fault-tolerant DRHC is based on two extensions of existingnwork for the case of large communication delays. The first contributionnis the development of a new DRHC approach that estimatesnthe trajectory of the neighboring vehicles for the tail of the predictionnhorizon, which would otherwise not be available due to the communicationndelay. In this approach, the tail of the cost function isnestimated by adding extra decision variables in the cost function. Anrelatively small amount of existing work has investigated thenimplementation issues associated with exchange of trajectory information,nbut so far no work has proposed a tail estimation process toncompensate for large delays. For instance, in [1–3], no prediction ornestimation for the trajectory of neighboring vehicles is performed,nand it is assumed that the neighboring vehicles remain at the lastndelayed states broadcasted by them. Such assumptions may yieldnpoor performance for large communication delays because thenconstant state vector is not a good estimation of a trajectory of statesnin general. Similar issues are also investigated in [4,5].