# US 2021/0276594 A1 · Vehicle trajectory prediction near or at traffic signal

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

- Number: US 2021/0276594 A1
- Document ID: US20210276594A1
- Status: Published application
- Kind: A1
- Application number: 17/150,690
- Filed: 2021-01-15
- Published: 2021-09-09
- Earliest claimed: 2020-01-15
- Assignees: The Regents of the University of Michigan
- Applicants: The Regents of the University of Michigan
- Inventors: Geunseob OH; Huei PENG
- Web page: https://patentgrep.com/patent/US20210276594A1

## CPC

- B60W 60/0027: {using trajectory prediction for other traffic participants}
- H04W 4/029: Location-based management or tracking services
- B60W 40/04: Traffic conditions
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G08G 1/0112: {from the vehicle, e.g. floating car data [FCD]}
- G08G 1/0116: {from roadside infrastructure, e.g. beacons}
- G08G 1/012: {from other sources than vehicle or roadside beacons, e.g. mobile networks}
- G08G 1/0129: {for creating historical data or processing based on historical data}
- G08G 1/095: Traffic lights
- G08G 1/096725: {where the received information generates an automatic action on the vehicle control}
- H04L 67/12: specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
- H04W 4/40: for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W 4/44: for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
- H04W 4/46: for vehicle-to-vehicle communication [V2V]
- B60W 2554/4042: Longitudinal speed
- B60W 2554/4045: Intention, e.g. lane change or imminent movement
- B60W 2554/80: Spatial relation or speed relative to objects
- B60W 2556/65: Data transmitted between vehicles
- 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
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- 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})
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A system and method for determining a predicted trajectory of a human-driven host vehicle as the human-driven host vehicle approaches a traffic signal. The method includes: obtaining a host vehicle-traffic light distance dx and a longitudinal host vehicle speed vx that are each taken when the human-driven host vehicle approaches the traffic signal; obtaining a traffic light signal phase Pt and an traffic light signal timing Tt; obtaining a time of day TOD; providing the host vehicle-traffic light distance dx, the longitudinal host vehicle speed vx, the traffic light signal phase Pt, the traffic light signal timing Tt, and the time of day TOD as input into an artificial intelligence (AI) vehicle trajectory prediction application, wherein the AI vehicle trajectory prediction application implements an AI vehicle trajectory prediction model; and determining the predicted trajectory of the human-driven host vehicle using the AI vehicle trajectory prediction application.

## Claims

### Claim 1

A method for determining a predicted trajectory of a human-driven host vehicle as the human-driven host vehicle approaches a traffic signal, wherein the method is carried out by one or more electronic controllers, and wherein the method comprises the steps of:
obtaining a host vehicle-traffic light distance dx and a longitudinal host vehicle speed vx that are each taken when the human-driven host vehicle approaches the traffic signal;
obtaining a traffic light signal phase Pt and a traffic light signal timing Tt, wherein the traffic light signal phase Pt represents a phase of the traffic signal taken when the human-driven host vehicle approaches the traffic signal, and wherein the traffic light signal timing Tt represents an amount of time elapsed since a last phase change of the traffic signal taken when the human-driven host vehicle approaches the traffic signal;
obtaining a time of day TOD;
providing the host vehicle-traffic light distance dx, the longitudinal host vehicle speed vx, the traffic light signal phase Pt, the traffic light signal timing Tt, and the time of day TOD as input into an artificial intelligence (AI) vehicle trajectory prediction application, wherein the AI vehicle trajectory prediction application implements an AI vehicle trajectory prediction model; and
determining the predicted trajectory of the human-driven host vehicle using the AI vehicle trajectory prediction application.

### Claim 2 (depends on 1)

The method of claim 1, wherein the method further includes obtaining a front vehicle state XFV, wherein the front vehicle state includes a front-host vehicle distance rt and a front-host vehicle speed {dot over (r)}t, and wherein the providing step further includes providing the front vehicle state XFV as input into the AI vehicle trajectory prediction application.

### Claim 3 (depends on 2)

The method of claim 2, wherein the front vehicle state XFV is obtained at the one or more electronic controllers based on front vehicle base information that is obtained at the front vehicle and then sent via vehicle-to-vehicle (V2V) communications to the one or more electronic controllers.

### Claim 4 (depends on 1)

The method of claim 1, wherein the traffic light signal phase Pt and the traffic light signal timing Tt are obtained from a traffic light control system that is present at an intersection where the traffic light is located.

### Claim 5 (depends on 1)

The method of claim 1, wherein the predicted trajectory is obtained at an autonomous vehicle that is approaching the traffic light and that is separate from the human-driven host vehicle.

### Claim 6 (depends on 5)

The method of claim 5, wherein the method is carried out at the autonomous vehicle as the human-driven host vehicle approaches the traffic light.

### Claim 7 (depends on 6)

The method of claim 6, wherein the autonomous vehicle obtains the traffic light signal phase Pt and the traffic light signal timing Tt from a traffic signal system located at an intersection where the traffic light is located.

### Claim 8 (depends on 7)

The method of claim 7, wherein the autonomous vehicle receives the traffic light signal phase Pt and the traffic light signal timing Tt via vehicle-to-infrastructure (V2I) communications from roadside equipment that is a part of the traffic signal system.

### Claim 9 (depends on 6)

The method of claim 6, wherein the autonomous vehicle receives the traffic light signal phase Pt and the traffic light signal timing Tt from a traffic signaling control system that is located remotely from the traffic light.

### Claim 10 (depends on 6)

The method of claim 6, wherein the host vehicle-traffic light distance dx, the longitudinal host vehicle speed vx are obtained at the autonomous vehicle via V2V communications with the host vehicle.

### Claim 11 (depends on 1)

The method of claim 1, wherein the vehicle-traffic light distance dx, the longitudinal host vehicle speed vx, the traffic light signal phase Pt, and the traffic light signal timing Tt are each associated with an associated time that is no more than a predetermined amount different than another one of the associated times.

### Claim 12 (depends on 1)

The method of claim 1, wherein the AI vehicle trajectory prediction model is or includes a neural network.

### Claim 13 (depends on 12)

The method of claim 12, wherein the AI vehicle trajectory prediction model is a deterministic or a model that predicts one or more most-probable trajectories.

### Claim 14 (depends on 12)

The method of claim 12, wherein the AI vehicle trajectory prediction model is a probabilistic model that returns a probability distribution of predicted trajectories, and wherein the predicted trajectory is obtained by sampling a trajectory from the probability distribution of predicted trajectories.

### Claim 15 (depends on 14)

The method of claim 14, wherein the neural network is a mixture density network.

### Claim 16 (depends on 12)

The method of claim 12, wherein the neural network is a deep neural network.

### Claim 17 (depends on 1)

The method of claim 1, wherein the method further includes the step of causing an autonomous vehicle to obtain the predicted trajectory of the human-driven vehicle, wherein the autonomous vehicle is configured to:
obtain the predicted trajectory of the human-driven vehicle, and
carry out an autonomous vehicle operation based on the predicted trajectory of the human-driven vehicle.

### Claim 18

A method for determining a predicted trajectory of a human-driven host vehicle as the human-driven host vehicle approaches a traffic signal, wherein the method is carried out by one or more electronic controllers, and wherein the method comprises the steps of:
obtaining a host vehicle-traffic light distance dx and a longitudinal host vehicle speed vx that are each taken when the human-driven host vehicle approaches the traffic signal, wherein the host vehicle-traffic light distance dx and the longitudinal host vehicle speed vx each have an associated time;
obtaining a front vehicle state XFV, wherein the front vehicle state includes a front-host vehicle distance rt and a front-host vehicle speed {dot over (r)}t;
receiving one or more wireless signals that indicate a traffic light signal phase Pt and a traffic light signal timing Tt, wherein the traffic light signal phase Pt represents a phase of the traffic signal taken when the human-driven host vehicle approaches the traffic signal, and wherein the traffic light signal timing Tt represents an amount of time elapsed since a last phase change of the traffic signal taken when the human-driven host vehicle approaches the traffic signal, wherein the traffic light signal phase Pt and the traffic light signal timing Tt each have an associated time, wherein the associated times of the host vehicle-traffic light distance dx, the longitudinal host vehicle speed vx, the traffic light signal phase Pt, the traffic light signal timing Tt, the front vehicle state includes a front-host vehicle distance rt, and the front-host vehicle speed {dot over (r)}t are within a maximum allowable time difference with respect to one another;
obtaining a time of day TOD;
providing the host vehicle-traffic light distance dx, the longitudinal host vehicle speed vx, the traffic light signal phase Pt, the traffic light signal timing Tt, the front vehicle state XFV, and the time of day TOD as input into an artificial intelligence (AI) vehicle trajectory prediction application, wherein the AI vehicle trajectory prediction application implements an AI vehicle trajectory prediction model, and wherein the AI vehicle trajectory prediction model is or includes a neural network; and
determining the predicted trajectory of the human-driven host vehicle using the AI vehicle trajectory prediction application.

### Claim 19 (depends on 18)

The method of claim 18, wherein the host vehicle-traffic light distance dx and the longitudinal host vehicle speed vx are both obtained at an autonomous vehicle through receiving one or more wireless signals from the human-driven host vehicle via vehicle-to-vehicle (V2V) communications.

### Claim 20 (depends on 18)

The method of claim 18, wherein the host vehicle-traffic light distance dx and the longitudinal host vehicle speed vx are both obtained at the autonomous vehicle through receiving one or more wireless signals from a remote server.