1.1Road geometry from a map
A satellite navigation receiver, inertial sensors, and wheel encoders estimate the vehicle pose. The controller retrieves nearby lane boundaries and transforms their coordinates into the vehicle frame.
Generated automatically. Not reviewed by a person. Not legal advice or an opinion.
Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.
A satellite navigation receiver, inertial sensors, and wheel encoders estimate the vehicle pose. The controller retrieves nearby lane boundaries and transforms their coordinates into the vehicle frame.
An image processor locates vehicles and pedestrians in successive frames from the lane camera. For objects on the road, it projects the bottom of each detection onto an estimated road plane and tracks the resulting positions over time.
A segmentation network labels road pixels by lane instance. The controller projects the pixel labels onto a ground-plane grid and identifies the driving lane from the cells beneath and immediately ahead of the vehicle.
The tracker projects each object's estimated footprint and position uncertainty onto ground-plane grid cells at successive future times. The controller checks which occupied cells overlap the area swept by each proposed vehicle path.
The controller samples the object footprints and the planned vehicle footprint at matching time steps. For each lane, it reports the track with the earliest predicted overlap and the time remaining until that overlap.
Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.
An autonomous yard tractor carries a forward camera and a lidar to monitor marked travel lanes between container rows. The image processor projects painted lane boundaries onto the yard surface, and the tracker transforms lidar detections into the same vehicle-relative coordinates. The controller assigns each tracked vehicle or obstacle to a lane by its ground position and reports the closest forward track in each lane. The tractor uses the distance in its travel lane to regulate speed as it approaches a loading position.
A forward color camera measures lane boundaries, and a separate synchronized stereo camera pair supplies object depth in place of radar. A stereo processor computes disparity and estimates each detected object's ground position from the depth points within its image mask. An extended Kalman filter estimates position and velocity across frames, and a calibrated rigid transform places each estimated position in the lane camera's ground coordinate frame. The lane processor assigns each track to a lane and sends the nearest forward track in each lane to the driving controller.
A forward camera runs at 30 frames per second, and front and front-corner radars update tracks at 20 hertz. The processor projects image-derived boundaries onto a ground plane and forms corridors for the host lane and up to two lanes on either side over a 120-meter forward range. Every 50 milliseconds, it compensates for vehicle motion, aligns the lane geometry and tracks to the same time, and projects each track onto the centerline of its assigned corridor. It reports the track with the smallest positive distance along each centerline to the driving controller.
A vehicle combines camera-derived lane corridors and radar tracks with a short-range vehicle-to-vehicle radio receiver. The processor transforms received vehicle positions into local coordinates and pairs each message with a radar track using position and speed gates. It assigns the radar tracks to the image-derived corridors and selects the nearest forward track in each lane. When the selected track in the host lane has a paired message, the cruise controller uses the reported acceleration as a feedforward input and regulates the gap with radar distance and relative speed.
A segmentation network trained on labeled road images and associated lane masks produces a separate image mask for each visible lane. The processor projects the masks onto a ground grid using the camera calibration and the measured vehicle pitch. For each radar track, it integrates the tracker’s ground-position probability distribution over the lane masks and assigns the track to the lane with the largest probability mass. It sends the closest forward track in each lane to the driving controller.
A camera processor fits lane boundaries on the ground plane, and a radar tracker supplies object position and lateral velocity at each update. A hidden Markov model maintains a lane state for each track, using signed distances to the lane boundaries as observations and assigning higher transition probabilities to adjacent lanes when lateral motion points toward their shared boundary. The processor uses the most probable current state to group tracks by lane and ranks each group by forward distance along the lane centerline. It outputs the nearest track in each group to the driving controller at every update.
A bumper-mounted frequency-modulated radar transmits chirps and samples the reflected signals. The processor estimates range and radial speed from the samples and uses phase differences across the receiving antennas to estimate bearing.
A camera processor identifies lane marking pixels and projects them onto an estimated road plane. It fits a cubic curve to each group of marking points and identifies the curves that bound the driving lane and an adjacent lane.
The controller transforms recent track positions into the ground-plane coordinates used for the image-derived lane boundaries. It sums the weights of the positions inside each lane, gives newer positions greater weight, and assigns the track to the lane with the highest sum.
Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.
The camera 10, for example, repeatedly captures an image of the vicinity of the host vehicle M periodically.A · US 2020/0312151 A1 · ¶ 36
The sensors 115 may further alternatively or additionally, for example, include camera sensor(s) 115, e.g. front view, side view, etc., providing images from an area surrounding the host vehicle 105.B · US 11,708,075 B2 · ¶ 37
The object recognition device 16 recognizes the position, type, speed, or the like of an object by performing a sensor fusion process on the results of detection performed by some or all of the camera 10, the radar device 12, and the viewfinder 14.A · US 2020/0312151 A1 · ¶ 39
For example, the sensors 115 may include Light Detection And Ranging (LIDAR) sensor(s) 115, etc., disposed on a top of the host vehicle 105, behind a host vehicle 105 front windshield, around the host vehicle 105, etc., that provide relative locations, sizes, and shapes of objects surrounding the host vehicle 105.B · US 11,708,075 B2 · ¶ 37
That is, the vehicle computer 110 can be programmed to receive sensor 115 data, typically, image data, from sensors 115 and to implement various image processing techniques to identify the host lane 205 and the target lane(s) 210. For example, lanes can be indicated by markings, e.g., painted lines on the road 200, and image recognition techniques, such as are known, can be executed by the vehicle computer 110 to identify the host lane 205.B · US 11,708,075 B2 · ¶ 52
The recognizer 130 is not limited to the road partition line, and may recognize the host lane and the adjacent lane by recognizing a driving boundary (road boundary) including a road partition line, a shoulder, a curbstone, a median strip, a guardrail, or the like.A · US 2020/0312151 A1 · ¶ 58
A group of dashed lane markings includes a plurality of markings spaced from each other along a length of a road and defining at least one boundary of a lane. Additionally, the vehicle computer 110 can determine the target lane(s) 210 on each side of the host lane 205 based on the number of groups of dashed lane markings on each side of the host vehicle 105 (e.g., a number of target lanes is equal to the number of groups of dashed lane markings).B · US 11,708,075 B2 · ¶ 52
Advantageously, and as described herein, the vehicle computer can predict a trajectory of a target vehicle entering the lane based on sensor data. The vehicle computer can determine whether the target vehicle will remain in the lane or pass through the lane based on the predicted trajectory.B · US 11,708,075 B2 · ¶ 10
For example, when in the cruise control mode, the vehicle computer 110 may detect the target vehicle 215 entering the host lane 205 by using image data to determine that the target vehicle 215 is crossing the lane markings defining a boundary between the host lane 205 and the target lane 210 from which the target vehicle 215 is departing, e.g., according to image processing techniques, as discussed above.B · US 11,708,075 B2 · ¶ 64
The following traveling event is an event of causing the host vehicle M to follow another vehicle (hereinafter referred to as a preceding vehicle X) which is present within a predetermined distance (within, for example, 100 [m]) in front of the host vehicle M and is closest to the host vehicle M.A · US 2020/0312151 A1 · ¶ 61
In an adaptive cruise control system, a vehicle computer can maintain or adjust the speed of the host vehicle in a lane based on a speed and position of a lead vehicle relative to the host vehicle. For example, the vehicle computer can actuate a braking component to reduce the speed of the host vehicle when the lead vehicle decelerates and/or is within a specified distance of the host vehicle.B · US 11,708,075 B2 · ¶ 9
An empty cell means only that no passage was reproduced there.
However, in the related art, even when a display was caused to display partition lines of a host lane and an adjacent lane, it was not possible to allow an occupant to understand whether the adjacent lane was a lane in which a lane change is possible in some cases.A · US 2020/0312151 A1 · ¶ 4
An aspect of the present invention is to provide a display control device, a display control method, and a storage medium that make it possible to allow an occupant to correctly understand whether or not to change a lane to an adjacent lane.A · US 2020/0312151 A1 · ¶ 5
patent.fyi, US 2026/0356393 A1, claim 1, first published 2026-11-20T01:05:31-05:00, https://patent.fyi/US20260356393/, SHA-256 69fbf6fbcbd7d65ea66d919eb1c720a6fa044f86f93a1835e5ba459141846e67