CPC G06V 10/7784 (2022.01) [G05D 1/0088 (2013.01); G05D 1/0246 (2013.01); G06T 7/60 (2013.01); G06T 7/90 (2017.01); G06V 20/20 (2022.01); G06V 20/582 (2022.01); G06T 2207/30252 (2013.01)] | 20 Claims |
1. A method of categorizing attributes of traffic signs for controlling an autonomous vehicle, the method comprising:
receiving, by one or more processors, sensor data captured by a perception system of the autonomous vehicle, the perception system including a plurality of sensors, and the sensor data including a plurality of traffic signs in surroundings of the autonomous vehicle in images detected by the sensors over a period of time;
categorizing, by the one or more processors, one or more of the traffic signs as being unidentifiable or unknown when a sign type of the one or more of the traffic signs cannot be identified with a confidence level that meets a particular confidence threshold level;
comparing, by the one or more processors, attributes of each of the categorized one or more traffic signs to attributes of known traffic signs;
determining, by the one or more processors, a sign type of one or more of the categorized one or more traffic signs based on results of the comparing; and
controlling, by the one or more processors, the autonomous vehicle based on the sign type of the one or more of the categorized one or more traffic signs.
|