# US 2023/0054974 A1 · Intersection Risk Indicator

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

- Number: US 2023/0054974 A1
- Document ID: US20230054974A1
- Status: Published application
- Kind: A1
- Application number: 17/408,907
- Filed: 2021-08-23
- Published: 2023-02-23
- Earliest claimed: 2021-08-23
- Assignees: Nissan North America, Inc.
- Applicants: Nissan North America, Inc.
- Inventors: Najamuddin Mirza Baig; Xin Yang
- Web page: https://patentgrep.com/patent/US20230054974A1

## CPC

- G08G 1/0141: {for traffic information dissemination}
- G08G 1/166: {for active traffic, e.g. moving vehicles, pedestrians, bikes}
- B60W 30/095: Predicting travel path or likelihood of collision
- B60W 30/18154: {Approaching an intersection}
- B60W 40/04: Traffic conditions
- B60W 40/09: Driving style or behaviour
- B60W 50/14: Means for informing the driver, warning the driver or prompting a driver intervention
- B60W 60/0015: {specially adapted for safety}
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 20/00: Machine learning
- G08G 1/0112: {from the vehicle, e.g. floating car data [FCD]}
- G08G 1/0116: {from roadside infrastructure, e.g. beacons}
- G08G 1/0133: {for classifying traffic situation}
- G08G 1/162: {event-triggered}
- G08G 1/163: {involving continuous checking}
- G08G 1/164: {Centralised systems, e.g. external to vehicles}
- B60W 2554/404: Characteristics
- G06V 20/58: Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- H04W 4/46: for vehicle-to-vehicle communication [V2V]
- 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})
- 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
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

A method and system for determining risk indicators for intersections. A method includes receiving, by a computing node from intersection proximate sensors at each intersection, intersection data. For each intersection, the method includes determining, by the computing node for each connected vehicle proximate to an intersection, a driver risk score based on driver distraction data from the intersection data for the intersection, determining, by the computing node, a near miss score based on the intersection data for the intersection, and assigning, by the computing node for the intersection, an intersection risk indicator level based on an exponential moving average of intersection risk indicator scores determined from driver risk scores and near miss scores. The method includes providing, by the computing node, intersection risk indicator levels to each connected vehicle to facilitate control decisions by each connected vehicle when approaching intersections.

## Claims

### Claim 1

A method for determining an intersection risk indicator, the method comprising:
receiving, by a computing node from intersection proximate sensors at each intersection, intersection data;
for each intersection:
determining, by the computing node for each connected vehicle proximate to an intersection, a driver risk score based on driver distraction data from the intersection data for the intersection;
determining, by the computing node, a near miss score based on the intersection data for the intersection;
assigning, by the computing node for the intersection, an intersection risk indicator level based on an exponential moving average of intersection risk indicator scores determined from driver risk scores and near miss scores; and
providing, by the computing node, intersection risk indicator levels to each connected vehicle to facilitate control decisions by each connected vehicle when approaching intersections.

### Claim 2 (depends on 1)

The method of claim 1, wherein the intersection proximate sensors includes at least connected vehicle sensors from connected vehicles proximate to each intersection, the determining the driver risk score further comprising:
clustering, using an unsupervised machine learning algorithm, the driver distraction data from each of the connected vehicles proximate to the intersection based on driving performance measures and driver physical measures; and
assigning a driver risk score for each risk cluster.

### Claim 3 (depends on 2)

The method of claim 2, wherein the intersection proximate sensors includes at least infrastructure sensors proximate each intersection, the determining the near miss score further comprising:
filtering the intersection data based on near miss criteria to detect near miss candidates;
classifying, using a machine learning algorithm, the near miss candidates as a near miss; and
assigning the near miss score based on number of near misses.

### Claim 4 (depends on 3)

The method of claim 3, wherein determining the intersection risk indicator scores further comprises:
determining a simple moving average of the driver risk scores and the near miss scores for a time period, wherein the exponential moving average is a weighted average of an intersection risk indicator score from a current time period and the simple moving average of a previous time period.

### Claim 5 (depends on 3)

The method of claim 3, wherein the near miss criteria includes one or more of braking time relative to obstacle, following distance relative to braking distance, obstacle distance relative to braking distance, vehicle deceleration, lane exceedance probability, or glance duration.

### Claim 6 (depends on 1)

The method of claim 1, wherein the intersection risk indicator levels are updated based on a defined period.

### Claim 7 (depends on 1)

The method of claim 1, wherein the intersection risk indicator levels are different for different times of day.

### Claim 8 (depends on 1)

The method of claim 1, the method further comprising:
providing, by the computing node, the intersection risk indicator levels to traffic management entities to facilitate traffic management at the intersections.

### Claim 9 (depends on 1)

The method of claim 1, the method further comprising:
providing, by the computing node, the intersection risk indicator levels to enable advanced driver-assistance systems on each connected vehicle to alert a driver based on the intersection risk indicator level.

### Claim 10 (depends on 1)

The method of claim 1, the method further comprising:
providing, by the computing node, the intersection risk indicator levels to each connected vehicle to populate a mapping database.

### Claim 11 (depends on 1)

The method of claim 1, wherein the computing node is one or more of a connected vehicle, an edge compute node, or a cloud computing node.

### Claim 12

A system, comprising:
a compute node;
intersection infrastructure sensors configured to send intersection infrastructure sensor data to the compute node; and
one or more connected vehicles configured to send intersection proximate data to the compute node,
wherein the compute node is configured to:
for each intersection that each of the one or more connected vehicles intersects, determine a driver risk score for each connected vehicle from driver distraction data in the intersection infrastructure sensor data and the intersection proximate data;
for each intersection, determine a near miss score based on the intersection infrastructure sensor data and the intersection proximate data for that intersection;
for each intersection, assign an intersection risk indicator level based on an exponential moving average of intersection risk indicator scores determined from driver risk scores and near miss scores for that intersection; and
for each of the one or more connected vehicles, provide intersection risk indicator levels to facilitate control decisions by each of the one or more connected vehicles.

### Claim 13 (depends on 12)

The system of claim 12, wherein the compute node is further configured to:
use an unsupervised machine learning algorithm to cluster the driver distraction data from each of the connected vehicles based on driving performance measures and driver physical measures and assign a driver risk score for each risk cluster.

### Claim 14 (depends on 12)

The system of claim 12, wherein the compute node is further configured to:
filter the intersection infrastructure sensor data and the intersection proximate data based on near miss criteria to detect near miss candidates; and
use a machine learning algorithm to classify the near miss candidates as a near miss and assign the near miss score based on number of near misses.

### Claim 15 (depends on 12)

The system of claim 12, wherein the compute node is further configured to:
determine a simple moving average of driver risk scores and near miss scores for a time period, wherein the exponential moving average is a weighted average of an intersection risk indicator score from a current time period and the simple moving average of a previous time period.

### Claim 16 (depends on 12)

The system of claim 12, wherein intersection risk indicator levels are updated based on a defined period and wherein the intersection risk indicator levels are different for different times of day.

### Claim 17 (depends on 12)

The system of claim 12, wherein the computing node is one or more of an edge compute node or a cloud computing node.

### Claim 18

An autonomous vehicle, comprising:
a sensor system having one or more vehicle sensors configured to send sensor data to a compute node;
one or more processors that execute computer-readable instructions that cause the one or more processors to:
receive, from the compute node, intersection risk indicator levels for intersections based on an exponential moving average of intersection risk indicator scores determined from driver risk scores and near miss scores using at least sensor data from the autonomous vehicle and other autonomous vehicles;
determine a control action for the autonomous vehicle to perform based on the intersection risk indicator level as the autonomous vehicle approaches an intersection; and
control the autonomous vehicle based on the control action.

### Claim 19 (depends on 18)

The autonomous vehicle of claim 18, wherein the control action is an alert to the user of the autonomous vehicle.

### Claim 20 (depends on 18)

The autonomous vehicle of claim 18, wherein the intersection risk indicator levels are updated based on a defined period and wherein the intersection risk indicator levels are different for different times of day.