US 12,403,933 B2
Determining a state of a vehicle on a road
Axel Beauvisage, Gothenburg (SE); Junsheng Fu, Nödinge (SE); Theodor Stenhammar, Gothenburg (SE); and David Bejmer, Gothenburg (SE)
Assigned to ZENSEACT AB, Gothenburg (SE)
Filed by ZENSEACT AB, Gothenburg (SE)
Filed on May 15, 2023, as Appl. No. 18/317,185.
Claims priority of application No. 22173553 (EP), filed on May 16, 2022.
Prior Publication US 2023/0365154 A1, Nov. 16, 2023
Int. Cl. B60W 50/00 (2006.01); B60W 60/00 (2020.01); G01C 21/30 (2006.01)
CPC B60W 60/001 (2020.02) [B60W 50/00 (2013.01); G01C 21/30 (2013.01); B60W 2050/0052 (2013.01); B60W 2420/00 (2013.01); B60W 2530/00 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method for determining a state of a vehicle on a road portion having two or more lanes, the vehicle comprising an Automated Driving System (ADS) feature, the method comprising:
obtaining map data associated with the road portion;
obtaining positioning data indicating a pose of the vehicle on the road;
obtaining sensor data from a sensor system of the vehicle;
initializing a plurality of filters for the road portion wherein one filter is initialized per lane of the road portion based on the obtained map data, the obtained positioning data, and the obtained sensor data, wherein each filter indicates an estimated state of the vehicle on the road portion;
associating one or more sensor data point(s) in the obtained sensor data to a corresponding map-element of the obtained map data;
determining one or more normalized similarity score(s) between the associated obtained map data and the obtained sensor data;
determining one or more multivariate time-series data based on the determined one or more normalized similarity score(s), wherein each multivariate time-series data is attributed to a corresponding initialized filter among the plurality of initialized filters; and
providing the one or more multivariate time-series data as input to a trained machine-learning algorithm; wherein the trained machine learning algorithm is configured for:
determining a confidence probability value for each initialized filter of the plurality of initialized filters by means of a probabilistic classifier;
selecting one of the initialized filters, by comparing the confidence probability values determined for each initialized filter in conjunction with one or more multi-objective optimized coefficient(s), each optimized coefficient being indicative of an optimization between a readiness performance indicator and an accuracy performance indicator for selecting a single initialized filter as an output of the machine learning algorithm indicative of a current state of the vehicle on the road portion; wherein the method further comprises:
controlling the ADS feature of the vehicle based on the selected initialized filter.