Evaluation of a Buried Cable Roadside Animal Detection System.

摘要

Animal-vehicle collisions (AVC) are a concern for departments of transportation as they translate into hundreds of human fatalities and billions of dollars in property damage each year. A recently published report states that the Virginia Department of Transportation (VDOT) currently spends over $4 million yearly to remove about 55,000 deer carcasses from its roadways. Currently, one of the most effective existing methods to reduce AVCs is the use of animal detection systems, which can detect animals near the roadway and alert approaching drivers accordingly. In order to reduce AVCs in Virginia, VDOT, in collaboration with the Virginia Tech Transportation Institute, proposed the evaluation of an innovative roadside animal detection system in naturalistic and controlled conditions. This type of system offers numerous apparent advantages over aboveground animal detection technologies when environmental interferences, such as precipitation and vegetation, and site-specific characteristics, such as topology, subsidence, and road curvature, are considered. The subject animal detection system (ADS), a 300-m-long buried dual-cable sensor, detects the crossing of large and medium-sized animals and provides data on their location along the length of the cable. The system has a central processor unit for control and communication and generates an invisible electromagnetic detection field around buried cables. When the detection field is perturbed, an alarm is declared and the location of the intrusion is determined. Target animals are detected based on their conductivity, size, and movement, with multiple simultaneous intrusions being detected during a crossing event. The system was installed and tested at a highly suitable site on the Virginia Smart Road where large wild animals, including deer and bear, are often observed in a roadside environment. This report describes the installation of the ADS, data collection and analysis methodology, evaluation of the system’s reliability and effectiveness, cost analysis, and implementation prospects. The system used continuous, all-weather and nighttime video surveillance to monitor animal movement and to gauge system detections, and potential non-detections of the ADS. Also, a communication link between the buried ADS and the Virginia Smart Road fiber optic network was established to allow operation and monitoring of the system from a dedicated server in the Virginia Smart Road Control Room. A performance verification of the network communication was successfully conducted through continuous data collection and transfer to a storage unit. Data were collected continuously for a period of 10 months that included winter, and then analyzed to determine overall detection performance of the system. Data analyses indicate that the ADS, if properly installed and calibrated, is capable of detecting animals such as deer and bear, and possibly smaller animals, such as fox and coyotes, with over 95% reliability. The ADS also performed well even when covered by 3 ft of snow. Moreover, the system was tested under various traffic conditions and no vehicle interferences were noted during the same monitoring period. The acquired data can be used to improve highway safety through driver warning systems installed along roadway sections where high wildlife activity has been observed. Additionally, this system may be integrated with the connected vehicle framework to provide advance, in-vehicle warnings to motorists approaching locations where animals have been detected in or near the roadway.