US 11,837,090 B2
System and methods of adaptive traffic rule-based decision making for autonomous driving
Biao Ma, Sunnyvale, CA (US); and Lior Tal, San Diego, CA (US)
Assigned to CYNGN, INC., Menlo Park, CA (US)
Filed by CYNGN, INC., Menlo Park, CA (US)
Filed on Aug. 5, 2022, as Appl. No. 17/817,945.
Claims priority of provisional application 63/366,738, filed on Jun. 21, 2022.
Claims priority of provisional application 63/229,850, filed on Aug. 5, 2021.
Claims priority of provisional application 63/229,852, filed on Aug. 5, 2021.
Claims priority of provisional application 63/229,856, filed on Aug. 5, 2021.
Prior Publication US 2023/0040017 A1, Feb. 9, 2023
Int. Cl. G08G 1/0967 (2006.01); G05D 1/00 (2006.01); G08G 1/01 (2006.01); B60W 60/00 (2020.01); B60W 40/04 (2006.01); B60W 50/00 (2006.01); B60W 30/09 (2012.01)
CPC G08G 1/096725 (2013.01) [B60W 30/09 (2013.01); B60W 40/04 (2013.01); B60W 50/0097 (2013.01); B60W 60/0015 (2020.02); B60W 60/0027 (2020.02); G05D 1/0088 (2013.01); G08G 1/0108 (2013.01); B60W 2552/53 (2020.02); B60W 2554/20 (2020.02); B60W 2554/404 (2020.02); B60W 2554/4026 (2020.02); B60W 2554/80 (2020.02); B60W 2555/60 (2020.02); G05D 2201/0213 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining input information relating to an environment in which an autonomous vehicle (AV) operates, the input information describing a geographical location in which the AV operates, a framework of traffic guidelines associated with the geographical location, and a configuration of roads in a vicinity of the AV;
setting up a traffic rule profile for the AV that specifies legal traffic practices corresponding to the geographical location in which the AV operates and the framework of traffic guidelines that the AV follows in the geographical location, the legal traffic practices differing according to a country, a state, or a city associated with the geographical location;
identifying a first traffic rule that applies to the AV based on the traffic rule profile and the configuration of the roads in the vicinity of the AV;
using a machine-learning process to prioritize traffic decisions made in relation to the identified traffic rule and the traffic rule profile and resolve internal conflicts between the traffic decisions;
adaptively prioritizing the traffic decisions based on a driving context so that a first traffic decision associated with a first traffic rule profile increases or decreases in priority in response to a selection or non-selection of other traffic rule profiles, wherein a separate priority ranking for conflicting traffic decisions is performed in response to two or more traffic decisions conflicting with each other;
determining a first driving decision that adheres to the identified first traffic rule and the legal traffic practices specified by the traffic rule profile;
sending a first instruction to a control system of the AV, the first instruction describing a first operation of the AV according to the first driving decision; and
controlling at least one vehicular system of the AV according to the first instruction such that the first operation accounts for the first driving decision.