US 12,094,273 B2
Systems and methods to improve the performance of anomaly detection
Seyhan Ucar, Mountain View, CA (US); Emrah Akin Sisbot, Mountain View, CA (US); and Kentaro Oguchi, Mountain View, CA (US)
Assigned to TOYOTA MOTOR ENGINEERING & MANUFACTURING NORTH AMERICA, INC., Plano, TX (US)
Filed by TOYOTA MOTOR ENGINEERING & MANUFACTURING NORTH AMERICA, INC., Plano, TX (US)
Filed on Feb. 24, 2022, as Appl. No. 17/680,050.
Prior Publication US 2023/0267782 A1, Aug. 24, 2023
Int. Cl. G07C 5/08 (2006.01); B60W 50/02 (2012.01)
CPC G07C 5/085 (2013.01) [B60W 50/0205 (2013.01); G07C 5/0816 (2013.01)] 32 Claims
OG exemplary drawing
 
1. A system to improve performance of anomaly detection in a vehicle, comprising:
a memory; and
one or more processors, wherein the one or more processors are configured to execute machine readable instructions stored in the memory for causing the one or more processors to:
monitor use of an anomaly detection mechanism, wherein the anomaly detection mechanism detects at least one of abnormal driving behavior or environment anomalies;
based on the monitoring, record metadata associated with the anomaly detection mechanism;
compare the metadata associated with the anomaly detection mechanism to driving behavior performed by the vehicle;
detect characteristic differences between the driving behavior performed by the vehicle and suggested vehicle operations in association with the anomaly detection,
wherein the characteristic differences are measured in excess of a threshold value for a location or a situation known to cause inaccurate performance responses of the anomaly detection mechanism, and
wherein the characteristic differences correspond with a decreased performance during monitored use of the anomaly detection mechanism;
based on the characteristic differences exceeding the threshold value, revise an anomaly detection rule for the location or the situation known to cause the inaccurate performance responses; and
reuse the anomaly detection mechanism with the revised anomaly detection rule.