# US 12,223,559 B1 · Natural language processing analysis of legal requirements > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,559 B1 - Document ID: US12223559B1 - Status: Granted - Kind: B1 - Application number: 17/561,549 - Filed: 2021-12-23 - Granted: 2025-02-11 - Earliest claimed: 2021-12-23 - Assignees: GM Cruise Holdings LLC - Applicants: GM Cruise Holdings LLC - Inventors: Kunal Mehta; Brent Valle; Ruixin Ge - Web page: https://patentgrep.com/patent/US12223559B1 ## CPC - G06Q 50/18: Legal services - G06F 9/451: Execution arrangements for user interfaces - G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking - G06F 40/30: Semantic analysis - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A computer-aided method of comparing legal requirements, including: comparing a first legal requirement from a first jurisdiction to a second legal requirement from a second jurisdiction, including applying a natural language processing (NLP) algorithm to the first and second legal requirements; computing a distance between the first and second legal requirements; determining from the distance that the second legal requirement belongs to a same subject matter category as the first legal requirement; and providing a user interface (UI) to display the first and second legal requirements to a user, with the computed distance. ## Claims ### Claim 1 A computer-aided method of comparing legal requirements, comprising: operating an autonomous vehicle to drive along a roadway; comparing a first legal requirement from a first jurisdiction to a second legal requirement from a second jurisdiction, comprising applying a natural language processing (NLP) algorithm to the first and second legal requirements; computing a distance between the first and second legal requirements; determining from the distance that the second legal requirement belongs to a same subject matter category as the first legal requirement; inferring a functional definition for the subject matter category; inferring a recommended parameterization for the functional definition; and instructing an autonomous vehicle controller of the autonomous vehicle to set a driving behavior in accordance with the recommended parameterization. ### Claim 2 (depends on 1) The computer-aided method of claim 1, further comprising providing a user interface (UI) to display the first and second legal requirements to a user, and displaying the computed distance. ### Claim 3 (depends on 1) The computer-aided method of claim 1, wherein the first and second legal requirements are statutes. ### Claim 4 (depends on 1) The computer-aided method of claim 1, wherein the first and second legal requirements are regulations or non-governmental rules. ### Claim 5 (depends on 2) The computer-aided method of claim 2, further comprising displaying one or more related laws distinct to the first jurisdiction or second jurisdiction. ### Claim 6 (depends on 1) The computer-aided method of claim 1, further comprising exporting linkages from a requirements tracking software, and associated the linkages with the first and second legal requirements. ### Claim 7 (depends on 1) The computer-aided method of claim 1, wherein computing the distance comprises computing a cosine or Jaccard distance. ### Claim 8 (depends on 1) The computer-aided method of claim 1, further comprising computing a distance for a plurality of legal requirements from a plurality of jurisdictions, determining that the plurality of legal requirements belong to the same subject matter category, and clustering the plurality of legal requirements by similarity. ### Claim 9 (depends on 8) The computer-aided method of claim 8, further comprising identifying a strictest rule for a cluster and generating parameters based on the strictest rule for the cluster, the instructing the autonomous vehicle controller of the autonomous vehicle to set the driving behavior is further in accordance with the generated parameters based on the strictest rule. ### Claim 10 (depends on 2) The computer-aided method of claim 2, further comprising providing the inferred functional definition via the UI in a human-readable form. ### Claim 11 (depends on 10) The computer-aided method of claim 10, further comprising providing the inferred recommended parameterization via the UI. ### Claim 12 (depends on 1) The computer-aided method of claim 1, wherein the NLP algorithm treats a first N words of the first and second legal requirements as a pseudo-sentence, wherein Nis approximately 400, and wherein the first N words include a title. ### Claim 13 A system for parameterizing requirements for an autonomous vehicle, comprising: a hardware platform comprising a processor circuit and a memory; instructions encoded within the memory to instruct the processor circuit to: receive a first compliance requirement from a first jurisdiction; receive a second compliance requirement from a second jurisdiction; apply a machine learning algorithm to the first and second compliance requirements to compute a distance between the first and second compliance requirements; generate an autonomous vehicle parameter based on the computed distance between the first and second compliance requirements; and instruct the autonomous vehicle to modify a driving behavior in accordance with the generated autonomous vehicle parameter; and the autonomous vehicle, the autonomous vehicle comprising an onboard computer configured to: navigate the autonomous vehicle to drive along a roadway, including performing the modified driving behavior in accordance with the generated autonomous vehicle parameter. ### Claim 14 (depends on 13) The computing apparatus of claim 13, wherein the machine learning algorithm comprises natural language processing (NLP). ### Claim 15 (depends on 14) The computing apparatus of claim 14, wherein the machine learning algorithm comprises a neural network. ### Claim 16 (depends on 13) The computing apparatus of claim 13, wherein the instructions are further to compute a distance for a plurality of compliance requirements from a plurality of jurisdictions, determining that the plurality of compliance requirements belong to a common subject matter category, and cluster the plurality of compliance requirements by similarity. ### Claim 17 (depends on 16) The computing apparatus of claim 16, wherein the instructions are further to generate a model statute from the plurality of compliance requirements. ### Claim 18 (depends on 16) The computing apparatus of claim 16, wherein the instructions are further to generate a test protocol for the common subject matter category. ### Claim 19 One or more tangible, non-transitory computer-readable storage media having stored thereon instructions executable on one or more processors, the instructions to instruct the one or more processors to: operate an autonomous vehicle to drive along a roadway; apply natural language processing (NLP) to a first legal requirement from a first jurisdiction; apply NLP to a second legal requirement from a second jurisdiction; compute a distance between the first and second legal requirements according to results of the NLP; assign the first and second legal requirements to a common subject matter category according to the distance; build a functional model for the common subject matter category based on the results of the NLP and the computed distance between the first and second legal requirements; and instruct an autonomous vehicle controller of the autonomous vehicle to modify a driving behavior in accordance with the functional model. ### Claim 20 (depends on 19) The one or more tangible, non-transitory computer-readable media of claim 19, wherein the first and second legal requirements are statutes.