High-Resolution Traffic Sensing with Probe Autonomous Vehicles: A Data-Driven Approach

Abstract

Recent decades have witnessed the breakthrough of autonomous vehicles (AVs), and the sensing capabilities of AVs have been dramatically improved. Various sensors installed on AVs will be collecting massive data and perceiving the surrounding traffic continuously. In fact, a fleet of AVs can serve as floating (or probe) sensors, which can be utilized to infer traffic information while cruising around the roadway networks. Unlike conventional traffic sensing methods relying on fixed location sensors or moving sensors that acquire only the information of their carrying vehicle, this paper leverages data from AVs carrying sensors for not only the information of the AVs, but also the characteristics of the surrounding traffic. A high-resolution data-driven traffic sensing framework is proposed, which estimates the fundamental traffic state characteristics, namely, flow, density and speed in high spatio-temporal resolutions and of each lane on a general road, and it is developed under different levels of AV perception capabilities and for any AV market penetration rate. Experimental results show that the proposed method achieves high accuracy even with a low AV market penetration rate. This study would help policymakers and private sectors (e.g., Waymo) to understand the values of massive data collected by AVs in traffic operation and management.

Motorway Measurement Campaign to Support R&D Activities in the Field of Automated Driving Technologies

“…The side zones of the tractors and the combination are covered at each side by wide angle short range radars (4 and 4 ) and front corner placed down looking fisheye cameras (8 and 8 ), where both of the fisheye cameras cover the vehicle front and the corresponding vehicle sides. The rear zone is covered at each side by rear looking cameras (7 and 7 ) and rear looking long range radars (5 and 5 ), each being installed to the mirror consoles of the vehicle and, finally, a rear looking long range radar (10), which is installed to the middle of the rear axle. If a trailer is connected, the rear middle long range radar (10) is looking out underneath the trailer to the ego lane.…”

“…It is a general problem that high precision map data are not available for automotive tests. Moreover, the sensing of the car and related processing/fusing algorithms cannot be validated against real-world data, as, usually, during the test measurement, there is no other external detection that records the car's motion on HD map [9,10]. Thus, it is a real problem to measure and validate the performance of a sensor fusion algorithm under real road conditions.…”

Low-Rank Hankel Tensor Completion for Traffic Speed Estimation

“…However, in practice, data is only available from a very small fraction of vehicles (e.g., penetration rate less than 5%). As a result, the traffic state information collected from sparse Lagrangian sensors will naturally be sparse in space and time [3,4].…”

Scenario-Mining for Level 4 Automated Vehicle Safety Assessment from Real Accident Situations in Urban Areas Using a Natural Language Process

“…However, it is challenging to generate representative test scenarios because there are many traffic situations on urban roads [ 7 ]. Various data sources such as traffic cameras and AVs sensor [ 8 ] datasets can be used to derive various situations on urban roads; however, such datasets are too extensive to analyze [ 9 , 10 ] and extracting only unsafe traffic conditions is more challenging [ 11 ]. Traffic accident data [ 12 ] are useful for extracting unsafe traffic conditions, critical objects, and behaviors that cause accidents.…”