US 12,450,436 B2
Language processing for a vehicle
Sasha Strelnikoff, Seattle, WA (US); Jiejun Xu, Diamond Bar, CA (US); and Alireza Esna Ashari Esfahani, Novi, MI (US)
Assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed by GM Global Technology Operations LLC, Detroit, MI (US)
Filed on Mar. 17, 2023, as Appl. No. 18/185,753.
Prior Publication US 2024/0311566 A1, Sep. 19, 2024
Int. Cl. G06F 40/289 (2020.01); B60K 35/10 (2024.01); B60K 35/22 (2024.01); G06F 40/205 (2020.01); G06V 20/58 (2022.01); G06V 30/262 (2022.01)
CPC G06F 40/289 (2020.01) [B60K 35/22 (2024.01); G06F 40/205 (2020.01); G06V 20/582 (2022.01); G06V 30/262 (2022.01); B60K 35/10 (2024.01); B60K 2360/148 (2024.01)] 17 Claims
OG exemplary drawing
 
1. A method for language processing for a vehicle, the method comprising:
receiving an input text, wherein the input text includes a plurality of words;
determining a rule-based action representation of the input text;
parsing the input text to produce a parsed text, wherein parsing the input text further comprises:
standardizing the input text to produce a standardized text, wherein each of a plurality of words of the standardized text is mapped to a corresponding one of the plurality of words of the input text;
determining a standardized part-of-speech tag of each of a plurality of forced words of the plurality of words of the standardized text using a lookup table;
determining a forced part-of-speech tag of each of a plurality of forced words of the plurality of words of the input text based on the standardized part-of-speech tag of the corresponding one of the plurality of forced words of the standardized text;
determining a first plurality of part-of-speech taggings of the input text using a machine learning algorithm, wherein each of the first plurality of part-of-speech taggings includes a predicted part-of-speech tag for each of the plurality of words of the input text, and wherein each of the first plurality of part-of-speech taggings has a confidence value; and
determining a part-of-speech tag of each of the plurality of words of the input text based at least in part on the confidence value of each of the first plurality of part-of-speech taggings and the plurality of forced words;
determining a model-based action representation of the parsed text;
determining a final action representation of the input text based at least in part on the rule-based action representation and the model-based action representation; and
taking an action based at least in part on the final action representation, wherein the action includes adjusting a routing parameter of an automated driving system of the vehicle.