Performance and Scaling of Collaborative Sensing and Networking for Automated Driving Applications

Abstract

A critical requirement for automated driving systems is enabling situational awareness in dynamically changing environments. To that end vehicles will be equipped with diverse sensors, e.g., LIDAR, cameras, mmWave radar, etc. Unfortunately the sensing 'coverage' is limited by environmental obstructions, e.g., other vehicles, buildings, people, objects etc. A possible solution is to adopt collaborative sensing amongst vehicles possibly assisted by infrastructure. This paper introduces new models and performance analysis for vehicular collaborative sensing and networking. In particular, coverage gains are quantified, as are their dependence on the penetration of vehicles participating in collaborative sensing. We also evaluate the associated communication loads in terms of the Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) capacity requirements and how these depend on penetration. We further explore how collaboration with sensing capable infrastructure improves sensing performance, as well as the benefits in utilizing spatiotemporal dynamics, e.g., collaborating with vehicles moving in the opposite direction. Collaborative sensing is shown to greatly improve sensing performance, e.g., improves coverage from 20% to 80% with a 20% penetration. In scenarios with limited penetration and high coverage requirements, infrastructure can be used to both help sense the environment and relay data. Once penetration is high enough, sensing vehicles provide good coverage and data traffic can be effectively 'offloaded' to V2V connectivity, making V2I resources available to support other in-car services.

Collective Perception: A Safety Perspective

“…The mentioned on-ramping scenario is a good example where the SAR may be more meaningful than the OAR, since the vehicle is interested in finding a suitable gap between the vehicles in the highway, and only knowing the position of some of the vehicles would not be sufficient. A SAR-like metric is made use of in [ 19 , 40 , 41 ].…”

“…It is composed of all arising latencies, the major ones being the just introduced update period, the sensor update rate, processing times and further communication delays [ 7 ]. Several papers have made use of the DOR [ 7 , 12 , 30 , 31 , 40 , 41 , 42 ], the update rate or period [ 6 , 7 , 11 , 12 , 32 , 35 ], the distance covered by the object between updates [ 6 , 7 , 32 ], and the AOI [ 6 , 7 ].…”

V2V Cooperative Sensing using Reinforcement Learning with Action Branching

“…For the RSU and vehicular agents training purposes, DQN [9] is selected as the algorithmic basis 3 . The detailed training algorithm is shown in Algorithm 1.…”

“…It can facilitate the exchange of sensory information between vehicles to enhance the perception of the surrounding environment beyond their sensing range; such process is called cooperative perception [2]. The advantages of cooperative perception are validated in [3] demonstrating that it greatly improves the sensing performance. Motivated by its potential, several standardization bodies are currently focusing their efforts towards formally defining the cooperative perception message (CPM), its contents and generation rate [2], [4].…”

Demonstrations of Cooperative Perception: Safety and Robustness in Connected and Automated Vehicle Operations

“…Most of the CP related use cases studied in the literature are safety related, including cooperative driving [ 5 , 16 ], cooperative advisory warnings [ 37 , 38 ], cooperative collision avoidance [ 4 , 13 , 39 ], intersection assistance [ 17 , 40 ], and vehicle misbehaviour detection [ 41 ], to name a few. It is presented in [ 42 ] quantitative comparison of V2V and V2I connectivity on improving sensing redundancy and collaborative sensing coverage for CAV applications. The work concludes that infrastructure support is crucial for safety related services such as CP, especially when the penetration rate of sensing vehicles is low.…”