Speed limits in autonomous vehicular networks due to communication constraints

Abstract

Autonomous vehicles need to be aware of other vehicles in their vicinity in order to avoid collisions and successfully perform their tasks. Such network awareness is ensured by exchanging location and control information over wireless radio channels. However, wireless interference constraints limit the number of messages that can be exchanged between the vehicles. In this paper, we study the impact of such communication constraints on maximum vehicle speed in dense autonomous vehicular networks. We define hazard rate to be the fraction of time a vehicle enters an `uncertainty region', i.e., a region where there is a positive probability of other vehicles being present due to lack of situational awareness. We show that the hazard rate follows a threshold behavior with respect to maximum speed v as the network density n increases to infinity. We show that for a 2D network the hazard rate tends to 1, if the maximum speed v decreases slower than n[superscript -3/2]; and tends to 0, if v decreases faster than n[superscript -3/2]. For the network hazard rate, which is the fraction of time any vehicle enters its uncertainty region, the threshold is n[superscript -2]. Finally, we extend these results to a 3D network and show that the thresholds for the 3D network are larger than in the 2D network.