# US 12,179,757 B2 · Method for filtering the driving route based on distributed Kalman filter for stabilization of route tracking of autonomous vehicles > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,179,757 B2 - Document ID: US12179757B2 - Status: Granted - Kind: B2 - Application number: 17/623,117 - Filed: 2020-11-30 - Published: 2023-10-05 - Granted: 2024-12-31 - Earliest claimed: 2020-11-25 - Assignees: Korea Electronics Technology Institute - Applicants: Korea Electronics Technology Institute - Inventors: Young Bo Shim; Kyoung Won Min; Haeng Seon Son; Seon Young Lee; Chang Gue Park - Web page: https://patentgrep.com/patent/US12179757B2 ## CPC - B60W 30/10: Path keeping {(cruise control for automatically following a preceding vehicle B60W30/165)} - B60W 60/001: {Planning or execution of driving tasks} - G06V 10/30: Noise filtering - G06V 20/588: {Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road} - B60W 2050/0052: {Filtering, filters} - B60W 2552/53: Road markings, e.g. lane marker or crosswalk - B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Provided is a method for filtering driving paths in order to stably track a path which is generated for various purposes, such as recognizing a lane and keeping or changing the lane, or avoiding an obstacle around a road. A driving path filtering system according to an embodiment of the present disclosure includes: a position recognition unit configured to recognize a position of a vehicle; a lane recognition unit configured to recognize a lane of a road; and a path generator configured to generate a path for the vehicle to travel on, based on a result of recognizing the position and a result of recognizing the lane, and to perform path noise filtering with respect to the generated path. Accordingly, a driving path can be stabilized by removing a path noise (error) by complementally filtering a lane recognition-based path generation method and a map-based path generation method according to a driving environment. ## Claims ### Claim 1 A driving path filtering system comprising: one or more processors comprising: a position recognition unit configured to recognize a position of a vehicle; a lane recognition unit configured to recognize a lane of a road; and a path generator configured to generate a map-based path for the vehicle to travel on, based on a result of recognizing the position of the vehicle, to generate a lane-based path based on a result of recognizing the lane, and to perform path noise filtering with respect to the generated map-based path or the generated lane-based path using a path noise removal filtering module wherein the one or more processors are configured to control a steering of the vehicle based on the generated map-based path or the generated lane-based path onto which the path noise filtering is performed, wherein the path noise removal filtering module comprises: a first local filter configured to filter the map-based path; a second local filter configured to filter the lane-based path; and a master filter configured to receive, as an input, a result of filtering through the first local filter or the second local filter, to correct an error associated with the map-based path or the lane-based path by removing a noise corresponding to the error associated with the map-based path or the lane-based path, and to feed a result of the removing the noise back to each of the first local filter and the second local filter. ### Claim 2 (depends on 1) The driving path filtering system of claim 1, wherein the path generator comprises: a map-based multi-purpose path generation module configured to receive the result of recognizing the position of the vehicle, and to generate the map-based path; a lane-based path generation module configured to receive the result of recognizing the lane, and to generate the lane-based path; a path selection module configured to select any one path from the map-based path and the lane-based path according to a predetermined selection condition; and the path noise removal filtering module configured to perform the path noise filtering with respect to the selected path. ### Claim 3 (depends on 2) The driving path filtering system of claim 2, wherein the path selection module is configured to, when a position recognition error of the map-based path is greater than or equal to a predetermined error threshold value, exclude the map-based path from the path selection, and, when performance of the lane-based path is less than or equal to a predetermined performance threshold value, exclude the lane-based path from the path selection even if the lane is recognized. ### Claim 4 (depends on 2) The driving path filtering system of claim 2, wherein a state vector xx of either the map-based path or the lane-based path is defined by Equation 1 presented below when a state k is formed of points that are spaced apart from one another by a predetermined distance l so as to define a set of points as one path: xk={p0,p1, . . . ,pn},∥pi−pj∥2=l Equation 1 ### Claim 5 (depends on 4) The driving path filtering system of claim 4, wherein the first local filter is configured to receive, as an input, the map-based path and to determine a covariance of the map-based path as a position recognition error, and wherein the second filter is configured to receive, as an input, the lane-based path and to determine a covariance of the lane-based path as lane recognition performance. ### Claim 6 (depends on 5) The driving path filtering system of claim 5, wherein the master filter is configured to receive an estimation value {circumflex over (x)}1, and the covariance P1 of the map-based path from the first local filter, and receive an estimation value {circumflex over (x)}2, and the covariance P2 of the lane-based path from the second local filter. ### Claim 7 A processor-implemented driving path filtering method comprising: recognizing, by a driving path filtering system, a position of a vehicle; recognizing, by the driving path filtering system, a lane of a road; generating, by the driving path filtering system, a map-based path for the vehicle to travel on, based on a result of recognizing the position of the vehicle; generating, by the driving path filtering system, a lane-based path based on a result of recognizing the lane of the road; performing, by the driving path filtering system, path noise filtering with respect to the generated map-based path or the generated lane-based path using a path noise removal filtering module; and controlling, by the driving path filtering system, a steering of the vehicle based on the generated map-based path or the generated lane-based path onto which the path noise filtering is performed, wherein the performing the path noise filtering comprises: filtering, using a first local filter, the map-based path; filtering, using a second local filter, the lane-based path; receiving, using a master filter, as an input, a result of filtering through the first local filter or the second local filter; correcting an error associated with the map-based path or the lane-based path by removing a noise corresponding to the error associated with the map-based path or the lane-based path; and providing a result of the removing the noise back to each of the first local filter and the second local filter. ### Claim 8 (depends on 7) The method of claim 7, further comprising: receiving the result of recognizing the position of the vehicle for generating the map-based path; receiving the result of recognizing the lane for generating the lane-based path; selecting any one path from the map-based path and the lane-based path according to a predetermined selection condition; and performing the path noise filtering with respect to the selected path. ### Claim 9 (depends on 8) The method of claim 8, wherein the selecting comprises: when a position recognition error of the map-based path is greater than or equal to a predetermined error threshold value, excluding the map-based path from the path selection, and when performance of the lane-based path is less than or equal to a predetermined performance threshold value, excluding the lane-based path from the path selection even if the lane is recognized. ### Claim 10 (depends on 8) The method of claim 8, wherein a state vector xk of either the map-based path or the lane-based path is defined by Equation 1 presented below when a state k is formed of points that are spaced apart from one another by a predetermined distance l so as to define a set of points as one path: xk={p0,p1, . . . ,pn},∥pi−pj∥2=l Equation 1 ### Claim 11 (depends on 10) The method of claim 10, wherein the first local filter is configured to receive, as an input, the map-based path and to determine a covariance of the map-based path as a position recognition error, and wherein the second filter is configured to receive, as an input, the lane-based path and to determine a covariance of the lane-based path as lane recognition performance. ### Claim 12 (depends on 11) The method of claim 11, wherein the master filter is configured to receive an estimation value {circumflex over (x)}1, and the covariance P1 of the map-based path from the first local filter, and receive an estimation value {circumflex over (x)}2, and the covariance P2 of the lane-based path from the second local filter.