Vehicular Path Prediction for Cooperative Driving Applications through Digital Map and Dynamic Vehicle Model Fusion

Abstract

This paper investigates the fusion of digital maps and dynamical models to accurately predict the future path of a vehicle. This path prediction is generated by a vehicle for sharing with other vehicles through wireless communication channels. This sharing is an enabler of Vehicle-to-Vehicle (V2V) cooperative systems. In particular, this work looks at cooper- ative safety and comfort systems, which benefit from accurate path predictions for collision avoidance and coordination with other vehicles. This work proposes that digital map information is valuable in giving a nominal estimate of the future path of the vehicle, and that this estimate can be augmented with real- time vehicle information, such as wheel speeds, yaw rates, and accelerations. Furthermore, dynamic short-term situations such as lane changes can be cleanly incorporated into the vehicular path prediction. Depending on the driving environment, the sensor quality, the map accuracy, and the map matching precision, this paper shows that this fusion has the potential to provide where-in-lane path predictions. I. INTRODUCTION Research into combining global navigation satellite sys- tems (GNSS) with wireless communication technologies is enabling future cooperative driving applications with benefits to safety, comfort, and mobility services. Comfort and mo- bility services, which look at reducing the driver's work load and increasing traffic flow respectively, likely will be the first applications of such wireless communications to production vehicles. These applications require fairly infrequent commu- nication updates and communication latency can be tolerated. Conversely, safety applications require high-frequency, low- latency communications that must contain precise vehicle positioning and orientation information. Although toughest on the communications requirements, it is safety applica- tions that can leverage the abundant amount of vehicle specific information in their message payloads. Whereas some cooperative mobility applications may be addressed by communication media (e.g., WiMAX) that is independent of the vehicle type or OEM-specific vehicle integration, safety applications must employ communication media (e.g., DSRC) with standardized message formats (e.g., SAE J2735) and security-layer definitions (i.e., IEEE 1609.2). This paper introduces a fusion technique for combining digital map data with vehicle specific measurements (e.g., CAN and GPS) to produce accurate short-time (i.e., 3- to 10-second) horizon path predictions. These path predictions incorporate dynamic vehicle models that are integrated over the time horizon to provide a continuous path prediction over the entire time horizon, and not just predicted vehicle positions at the end of the time horizon. The purpose of