# US 2021/0061269 A1 · Method of handling occlusions at intersections in operation of autonomous vehicle

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0061269 A1
- Document ID: US20210061269A1
- Status: Published application
- Kind: A1
- Application number: 16/556,321
- Filed: 2019-08-30
- Published: 2021-03-04
- Earliest claimed: 2019-08-30
- Assignees: Argo AI, LLC
- Applicants: Argo AI, LLC
- Inventors: Thomas Petroff; Neal Seegmiller; Albert Costa; Christopher Cunningham; G. Peter K. Carr; Sameer Bardapurkar
- Web page: https://patentgrep.com/patent/US20210061269A1

## CPC

- B60W 30/0956: {the prediction being responsive to traffic or environmental parameters}
- B60W 30/09: Taking automatic action to avoid collision, e.g. braking and steering
- B60W 30/181: {Preparing for stopping}
- B60W 30/18154: {Approaching an intersection}
- B60W 30/18159: {Traversing an intersection}
- B60W 40/04: Traffic conditions
- B60W 60/0027: {using trajectory prediction for other traffic participants}
- G05D 1/0055: {with safety arrangements}
- G05D 1/0088: {characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours (using knowledge based models G06N5/00)}
- G08G 1/166: {for active traffic, e.g. moving vehicles, pedestrians, bikes}
- B60W 2552/53: Road markings, e.g. lane marker or crosswalk
- B60W 2554/00: Input parameters relating to objects
- B60W 2554/20: Static objects
- B60W 2554/4048: Field of view, e.g. obstructed view or direction of gaze
- B60W 2555/60: Traffic rules, e.g. speed limits or right of way
- B60W 2720/10: Longitudinal speed
- 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
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93})

## Abstract

An autonomous vehicle navigates an intersection in which occlusions block the vehicle's ability to detect moving objects. The vehicle handles this by generating a phantom obstacle behind the occlusion. The vehicle will predict the speed of the phantom obstacle and use the predicted speed to assess whether the phantom obstacle may collide with the vehicle. If a collision is a risk, the vehicle will slow or stop until it confirms that either (a) the phantom obstacle is not a real obstacle or (b) the vehicle can proceed at a speed that avoids the collision. To determine which occlusions shield real objects, the system may use a rasterized visibility grid of the area to identify occlusions that may accommodate the object.

## Claims

### Claim 1

A method of navigating an autonomous vehicle through an intersection, the method comprising:
by a processor that is controlling operation of one or more systems of a vehicle having a plurality of sensors that are configured to detect objects in an environment of the vehicle:
causing the vehicle to move in a path of travel in a first lane at a speed of travel;
detecting that an intersection with a second lane is positioned ahead of the autonomous vehicle in the path of travel;
determining a conflict region in the path of travel at the intersection;
before the vehicle reaches the intersection, analyzing data collected by the sensors to determine whether the data indicates that an occlusion exists that occludes the sensors' ability to detect whether an object exists in the second lane within a target distance of the intersection;
if the processor determines that no occlusion exists, causing the vehicle to proceed through the intersection while monitoring the conflicting lane for real obstacles;
if the processor determines that an occlusion exists, causing the vehicle to monitor for potential conflicting obstacles in the second lane by:
generating a phantom obstacle that is positioned behind the occlusion and that is moving toward the intersection in the second lane,
generating a predicted speed of the phantom obstacle, and
using the predicted speed of the phantom obstacle to determine whether the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region;
if the phantom obstacle is not expected to reach the conflict region before the vehicle will pass through the conflict region, causing the vehicle to continue to move in the path of travel and into the intersection; and
if the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action before permitting the vehicle to enter the intersection.

### Claim 2 (depends on 1)

The method of claim 1, wherein taking the safety-related action comprises:
reducing the speed of the travel of the vehicle;
analyzing new data collected by the sensors to determine whether the data indicates that the occlusion still occludes the sensors' ability to detect whether an object exists in the second lane within a target distance of the intersection; and
continuing to monitor for potential conflicting obstacles in the second lane until the earlier of (a) the vehicle reaches the conflict region, or (b) the processor determines that the phantom obstacle is a real obstacle.

### Claim 3 (depends on 1)

The method of claim 1, wherein taking the safety-related action comprises:
determining that the vehicle can safely pass through the intersection if the vehicle accelerates to an increased speed that is within a speed limit of the path of travel; and
causing the vehicle to pass through the intersection at the increased speed.

### Claim 4 (depends on 1)

The method of claim 1, wherein monitoring the conflicting lane for real obstacles comprises:
analyzing additional data collected by the sensors to search for real obstacles in the conflicting lane;
upon identifying that a real obstacle is approaching the intersection in the conflicting lane, determining whether the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region; and
if the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action before permitting the vehicle to enter the intersection.

### Claim 5 (depends on 1)

The method of claim 1 further comprising, by the processor:
if the processor determines that the phantom obstacle is a real obstacle before the vehicle clears the conflict region:
determining whether the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, and
if the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action.

### Claim 6 (depends on 1)

The method of claim 1 further comprising, by the processor:
determining a stop line position in the path of travel between the vehicle and the intersection;
when the autonomous vehicle reaches the stop line, causing the vehicle to either stop or reduce the speed of travel; and
when the vehicle moves between the stop line and the intersection, limiting the speed of travel to a threshold speed.

### Claim 7 (depends on 1)

The method of claim 1, wherein generating the phantom obstacle comprises determining whether an occluded object may exist behind the occlusion by:
accessing a rasterized visibility grid of the environment;
identifying a plurality of sampling points in the second lane on the rasterized visibility grid;
for each of the sampling points, applying a mask to an area around the sampling point; and
determining whether an occluded object may exist at the sampling point by determining whether all pixels within the mask are occluded.

### Claim 8 (depends on 7)

The method of claim 7, wherein generating the phantom obstacle further comprises generating the phantom obstacle at any location at which the system determines that an occluded object may exist.

### Claim 9 (depends on 7)

The method of claim 7 further comprising, after applying the mask to the area and before determining whether an occluded object may exist at the sampling point, convolving the mask with the rasterized visibility grid with the mask centered at the sampling point.

### Claim 10

A navigation system for an autonomous vehicle, the system comprising:
a processor;
a plurality of sensors that are configured to detect objects in an environment of a vehicle; and
a computer-readable medium containing programming instructions that are configured to instruct the processor to navigate the vehicle through an intersection by:
causing the vehicle to move in a path of travel in a first lane at a speed of travel,
detecting that an intersection with a second lane is positioned ahead of the autonomous vehicle in the path of travel,
determining a conflict region in the path of travel at the intersection, before the vehicle reaches the intersection, analyzing data collected by the sensors to determine whether the data indicates that an occlusion exists that occludes the sensors' ability to detect whether an object exists in the second lane within a target distance of the intersection,
if the processor determines that no occlusion exists, causing the vehicle to proceed through the intersection while monitoring the second lane for real obstacles,
if the processor determines that an occlusion exists, causing the vehicle to monitor the second lane for potential conflicting obstacles by:
generating a phantom obstacle that is positioned behind the occlusion and that is moving toward the intersection in the second lane;
generating a predicted speed of the phantom obstacle; and
using the predicted speed of the phantom obstacle to determine whether the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region,
if the phantom obstacle is not expected to reach the conflict region before the vehicle will pass through the conflict region, causing the vehicle to continue to move in the path of travel and into the intersection, and
if the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action before permitting the vehicle to enter the intersection.

### Claim 11 (depends on 10)

The system of claim 10, wherein the instructions to take the safety-related action comprise instructions to:
reduce the speed of the travel of the vehicle;
analyze new data collected by the sensors to determine whether the data indicates that the occlusion still occludes the sensors' ability to detect whether an object exists in the second lane within a target distance of the intersection; and
continuing to monitor for potential conflicting obstacles in the conflicting lane until the earlier of (a) the vehicle reaches the conflict region, or (b) the processor determines that the phantom obstacle is a real obstacle.

### Claim 12 (depends on 10)

The system of claim 10, wherein the instructions to take the safety-related action comprise instructions to:
determine that the vehicle can safely pass through the intersection if the vehicle accelerates to an increased speed that is within a speed limit of the path of travel; and
cause the vehicle to pass through the intersection at the increased speed.

### Claim 13 (depends on 11)

The system of claim 11 wherein the instructions that are configured to instruct the processor to monitor the second lane for real obstacles comprise instructions to:
analyze additional data collected by the sensors to search for real obstacles in the second lane;
upon identifying that a real obstacle is approaching the intersection in the second lane, determine whether the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region; and
if the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, take a safety-related action before permitting the vehicle to enter the intersection.

### Claim 14 (depends on 10)

The system of claim 10 further comprising additional instructions that are configured to instruct the processor to:
if the processor determines that the phantom obstacle is a real obstacle before the vehicle clears the conflict region:
determining whether the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, and
if the real obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action.

### Claim 15 (depends on 10)

The system of claim 10 further comprising additional instructions that are configured to instruct the processor to:
determine a stop line position in the path of travel between the vehicle and the intersection;
when the vehicle reaches the stop line, cause the vehicle to either stop or reduce the speed of travel; and
when the vehicle moves between the stop line and the intersection, limit the speed of travel to a threshold speed.

### Claim 16 (depends on 10)

The system of claim 10, wherein the instructions to generate the phantom obstacle comprise instructions to determine whether an occluded object may exist behind the occlusion by:
accessing a rasterized visibility grid of the environment;
identifying a plurality of sampling points in the second lane on the rasterized visibility grid;
for each of the sampling points, applying a mask to an area around the sampling point; and
determining whether an occluded object may exist at the sampling point by determining whether all pixels within the mask are occluded.

### Claim 17 (depends on 16)

The system of claim 16, wherein the instruction to generate the phantom obstacle further comprise instructions to generate the phantom obstacle at any location where the system determines that an occluded object may exist.

### Claim 18 (depends on 16)

The system of claim 16 further comprising instructions to, after applying the mask to the area and before determining whether an occluded object may exist at the sampling point, convolve the mask with the rasterized visibility grid with the mask centered at the sampling point.

### Claim 19

A computer program product for navigating a vehicle through an intersection, the product comprising programming instructions stored on a non-transitory computer-readable medium, wherein the instructions are configured to instruct a processor of a vehicle to navigate the vehicle through an intersection by:
causing the vehicle to move in a path of travel in a first lane at a speed of travel,
detecting that an intersection with a second lane is positioned ahead of the autonomous vehicle in the path of travel;
determining a conflict region in the path of travel at the intersection;
before the vehicle reaches the intersection, analyzing data collected by one or more sensors of the vehicle to determine whether the data indicates that an occlusion exists that occludes the one or more sensors' ability to detect whether an object exists in the second lane within a target distance of the intersection;
if the processor determines that no occlusion exists, causing the vehicle to proceed through the intersection while monitoring the second lane for real obstacles;
if the processor determines that an occlusion exists, causing the vehicle to monitor the second lane for potential conflicting obstacles by:
generating a phantom obstacle that is positioned behind the occlusion and that is moving toward the intersection in the second lane,
generating a predicted speed of the phantom obstacle, and
using the predicted speed of the phantom obstacle to determine whether the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region;
if the phantom obstacle is not expected to reach the conflict region before the vehicle will pass through the conflict region, causing the vehicle to continue to move in the path of travel and into the intersection; and
if the phantom obstacle is expected to reach the conflict region before the vehicle will pass through the conflict region, taking a safety-related action before permitting the vehicle to enter the intersection.

### Claim 20 (depends on 19)

The computer program product of claim 19, wherein the instructions to take the safety-related action comprise instructions to:
if the vehicle can safely pass through the intersection if the vehicle accelerates to an increased speed that is within a speed limit of the path of travel, cause the vehicle to pass through the intersection at the increased speed; and
if the vehicle reaches the conflict region before the processor determines that the phantom obstacle is a real obstacle, cause the vehicle to pass through the conflict region.
