US-12620238-B2 - Systems and methods for map-based real-world modeling

US12620238B2US 12620238 B2US12620238 B2US 12620238B2US-12620238-B2

Abstract

A system for correlating drive information from multiple road segments is disclosed. In one embodiment, the system includes memory and a processor configured to receive drive information from vehicles that traversed a first road segment and vehicles that traversed a second road segment. The processor is configured to correlate the drive information from the vehicles to provide a first road model segment representative of the first road segment and a second road model segment representative of the second road segment. The processor correlates the first road model segment with the second road model segment to provide a correlated road segment model if a drivable distance between a first point associated with the first road segment and a second point associated with the second road segment is less than or equal to a predetermined distance threshold, and stores the correlated road segment model as part of a sparse navigational map.

Inventors

Moshe SHENFELD
Ruth Chapman
Iddo Hanniel
Yael HACOHEN
Ishay GOLINSKY

Assignees

MOBILEYE VISION TECHNOLOGIES LTD.

Dates

Publication Date: 20260505
Application Date: 20230222

Claims (20)

1 . A system for correlating drive information from multiple road segments, the system comprising: at least one processor comprising circuitry and a memory, wherein the memory includes instructions that when executed by the circuitry cause the at least one processor to: receive drive information from each of a first plurality of vehicles that traversed an upper road segment on an overpass, wherein the drive information from each of the first plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the first plurality of vehicles traversed the upper road segment; correlate the drive information from each of the first plurality of vehicles to provide a first road model segment representative of the upper road segment, the first road model segment including at least a first point associated with the upper road segment; receive drive information from each of a second plurality of vehicles that traversed a lower road segment under the overpass, wherein the drive information from each of the second plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the second plurality of vehicles traversed the lower road segment; correlate the drive information from each of the second plurality of vehicles to provide a second road model segment representative of the lower road segment, the second road model segment including at least a second point associated with the lower road segment; determine that the first point and the second point meet a physical proximity criteria for correlating the lower road model segment and the upper road model segment; obtain drivable distance data indicative of a drivable distance between the first point and the second point, the drivable distance data being determined based on data retrieved from at least one road navigational model; determine that the drivable distance exceeds a drivable distance threshold; store the first road model segment and the second road model segment separately as part of a navigational map based on at least a result of the determining that the drivable distance exceeds the drivable distance threshold; and make the navigational map available to at least one host vehicle for use in navigation along the first road segment and the second road segment.
2 . The system of claim 1 , wherein the drive information received from the first plurality of vehicles and the drive information received from the second plurality of vehicles also include indications of objects detected through analysis of capture image frames and at least one positional indicator associated with each of the detected objects.
3 . The system of claim 2 , wherein the indications of the detected objects identify a non-semantic feature type.
4 . The system of claim 3 , wherein the non-semantic feature type includes at least one of a traffic sign, a traffic light, a lane marking, a manhole cover, a road edge, or a curb.
5 . The system of claim 2 , wherein the at least one positional indicator includes an X-Y location from a captured image frame.
6 . The system of claim 2 , wherein the at least one positional indicator includes a 3D feature point determined based on a structure in motion calculation.
7 . The system of claim 1 , wherein correlation of the drive information from each of the first plurality of vehicles includes aggregation and alignment of actual vehicle trajectories to determine a target trajectory for at least one lane of travel associated with the first road segment.
8 . The system of claim 7 , wherein the target trajectory is represented as a 3D spline.
9 . The system of claim 1 , wherein correlation of the drive information from each of the first plurality of vehicles includes aggregation of position indicators associated with objects detected based on analysis of captured image frames representative of the first road segment, and wherein the aggregation of position indicators is used to determine at least one refined position associated with each of the detected objects.
10 . The system of claim 1 , wherein correlation of the first road model segment and the second road model segment includes aligning target trajectories associated with the first and second road segment models.
11 . The system of claim 1 , wherein correlation of the first road model segment and the second road model segment includes meshing together representations of road surfaces associated with the first road model segment and the second road model segment.
12 . The system of claim 11 , wherein the meshing includes application of a triangulation algorithm.
13 . A method for correlating drive information from multiple road segments, the method comprising: receiving drive information from each of a first plurality of vehicles that traversed an upper road segment on an overpass, wherein the drive information from each of the first plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the first plurality of vehicles traversed the upper road segment; correlating the drive information from each of the first plurality of vehicles to provide a first road model segment representative of the upper road segment, the first road model segment including at least a first point associated with the upper road segment; receiving drive information from each of a second plurality of vehicles that traversed a lower road segment under the overpass, wherein the drive information from each of the second plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the second plurality of vehicles traversed the lower road segment; correlating the drive information from each of the second plurality of vehicles to provide a second road model segment representative of the lower road segment, the second road model segment including at least a second point associated with the lower road segment; determining that the first point and the second point meet a physical proximity criteria for correlating the lower road model segment and the upper road model segment; obtaining drivable distance data indicative of a drivable distance between the first point and the second point, the drivable distance data being determined based on data retrieved from at least one road navigational model; determining that the drivable distance exceeds a drivable distance threshold; storing the first road model segment and the second road model segment separately as part of a navigational map based on at least a result of the determining that the drivable distance exceeds the drivable distance threshold; and making the sparse navigational map available to at least one host vehicle for use in navigation along the first road segment and the second road segment.
14 . A non-transitory computer readable medium containing instructions that when executed by at least one processor, cause the at least one processor to perform a method for correlating drive information from multiple road segments, the method comprising: receiving drive information from each of a first plurality of vehicles that traversed an upper road segment on an overpass, wherein the drive information from each of the first plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the first plurality of vehicles traversed the upper road segment; correlating the drive information from each of the first plurality of vehicles to provide a first road model segment representative of the upper road segment, the first road model segment including at least a first point associated with the upper road segment; receiving drive information from each of a second plurality of vehicles that traversed a lower road segment under the overpass, wherein the drive information from each of the second plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the second plurality of vehicles traversed the lower road segment; correlating the drive information from each of the second plurality of vehicles to provide a second road model segment representative of the lower road segment, the second road model segment including at least a second point associated with the lower road segment; determining that the first point and the second point meet a physical proximity criteria for correlating the lower road model segment and the upper road model segment; obtaining drivable distance data indicative of a drivable distance between the first point and the second point, the drivable distance data being determined based on data retrieved from at least one road navigational model; determining that the drivable distance exceeds a drivable distance threshold; storing the first road model segment and the second road model segment separately as part of a navigational map based on at least a result of the determining that the drivable distance exceeds the drivable distance threshold; and making the navigational map available to at least one host vehicle for use in navigation along the first road segment and the second road segment.
15 . The system of claim 1 , wherein the drivable distance data represents a distance a hypothetical vehicle would travel between the first point and the second point, assuming the hypothetical vehicle abides by at least one of a traffic law, regulation, guideline, or best practice.
16 . The system of claim 1 , wherein execution of the instructions included in the memory further cause the at least one processor to: receive drive information from each of a third plurality of vehicles that traversed the upper road segment on an overpass, wherein the drive information from each of the third plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the third plurality of vehicles traversed the upper road segment; correlate the drive information from each of the third plurality of vehicles to provide a third road model segment representative of the upper road segment, the third road model segment including at least a third point associated with the upper road segment; obtain second drivable distance data indicative of a drivable distance between the first point and the third point, the second drivable distance data being determined based on data retrieved from the at least one road navigational model; correlate the first road model segment with the third road model segment to provide a correlated road segment model, wherein the first road model segment is correlated with the third road model based on at least the second drivable distance and a physical distance between the first point and the third point; and store the correlated road segment model as part of the navigational map.
17 . The system of claim 1 , wherein the drivable distance threshold is greater than a physical distance between the first point and the second point by a factor of at least 1.5.
18 . The system of claim 1 , wherein the drivable distance threshold is 30 meters or less.
19 . The system of claim 1 , wherein the distance threshold is 150% of the physical distance separating the first point and the second point.
20 . The system of claim 1 , wherein the drivable distance threshold is greater than a physical distance separating the first point and the second point by a predetermined ratio.

Description

CROSS REFERENCE TO RELATED APPLICATION This application is a continuation of PCT International Application No. PCT/US2021/048419, filed Aug. 31, 2021, which claims the benefit of priority of U.S. Provisional Application No. 63/072,597, filed Aug. 31, 2020. The foregoing applications are incorporated herein by reference in their entirety. BACKGROUND Technical Field The present disclosure relates generally to vehicle navigation. Background Information As technology continues to advance, the goal of a fully autonomous vehicle that is capable of navigating on roadways is on the horizon. Autonomous vehicles may need to take into account a variety of factors and make appropriate decisions based on those factors to safely and accurately reach an intended destination. For example, an autonomous vehicle may need to process and interpret visual information (e.g., information captured from a camera) and may also use information obtained from other sources (e.g., from a GPS device, a speed sensor, an accelerometer, a suspension sensor, etc.). At the same time, in order to navigate to a destination, an autonomous vehicle may also need to identify its location within a particular roadway (e.g., a specific lane within a multi-lane road), navigate alongside other vehicles, avoid obstacles and pedestrians, observe traffic signals and signs, and travel from one road to another road at appropriate intersections or interchanges. Harnessing and interpreting vast volumes of information collected by an autonomous vehicle as the vehicle travels to its destination poses a multitude of design challenges. The sheer quantity of data (e.g., captured image data, map data, GPS data, sensor data, etc.) that an autonomous vehicle may need to analyze, access, and/or store poses challenges that can in fact limit or even adversely affect autonomous navigation. Furthermore, if an autonomous vehicle relies on traditional mapping technology to navigate, the sheer volume of data needed to store and update the map poses daunting challenges. SUMMARY Embodiments consistent with the present disclosure provide systems and methods for vehicle navigation. In an embodiment, a system for correlating drive information from multiple road segments may include at least one processor comprising circuitry and a memory. The memory may include instructions that when executed by the circuitry cause the at least one processor to receive drive information from each of a first plurality of vehicles that traversed a first road segment, wherein the drive information from each of the first plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the first plurality of vehicles traversed the first road segment; and correlate the drive information from each of the first plurality of vehicles to provide a first road model segment representative of the first road segment. The at least one processor may further receive drive information from each of a second plurality of vehicles that traversed a second road segment, wherein the drive information from each of the second plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the second plurality of vehicles traversed the second road segment; and correlate the drive information from each of the second plurality of vehicles to provide a second road model segment representative of the second road segment. The at least one processor may further correlate the first road model segment with the second road model segment to provide a correlated road segment model if a drivable distance between a first point associated with the first road segment and a second point associated with the second road segment is less than or equal to a predetermined distance threshold; and store the correlated road segment model as part of a sparse navigational map for use in navigation of vehicles along the first road segment and the second road segment. In an embodiment, a method for correlating drive information from multiple road segments may include receiving drive information from each of a first plurality of vehicles that traversed a first road segment, wherein the drive information from each of the first plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the first plurality of vehicles traversed the first road segment; and correlating the drive information from each of the first plurality of vehicles to provide a first road model segment representative of the first road segment. The method may further include receiving drive information from each of a second plurality of vehicles that traversed a second road segment, wherein the drive information from each of the second plurality of vehicles includes at least an indication of an actual trajectory followed while a particular vehicle among the second plurality of vehicles traversed the second road segment;