Method and system for adaptively controlling object spacing

Patent metadata

Field	Value
Publication number	US-12459509-B2
Application number	US-202218070099-A
Country	US
Kind code	B2
Filing date	Nov 28, 2022
Priority date	Apr 27, 2018
Publication date	Nov 4, 2025
Grant date	Nov 4, 2025

Abstract

A method or system for adaptive vehicle spacing, including determining a current state of a vehicle based on sensor data captured by sensors of the vehicle; for each possible action in a set of possible actions: (i) predicting based on the current vehicle state a future state for the vehicle, and (ii) predicting, based on the current vehicle state a first zone future safety value corresponding to a first safety zone of the vehicle; and selecting, based on the predicted future states and first zone future safety values for each of the possible actions in the set, a vehicle action.

First claim

The invention claimed is:

1 . A computer-implemented method for adaptively controlling spacing between a vehicle and a moving object in an operating environment of the vehicle, the method comprising:
training a machine learning model using reinforcement learning to select vehicle responses based on predicted vehicle states and safety values, wherein the machine learning model is trained using a general value function (GVF) framework; determining a current vehicle state of the vehicle and a current object state of the moving object based on sensor data captured by sensors of the vehicle; predicting, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in a set of multiple alternative responses, a future state for the vehicle corresponding to the response; predicting, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in the set of multiple alternative responses, a first zone future safety value corresponding to a first safety zone of the vehicle, the first zone future safety value indicating a safety level of the vehicle for the first safety zone if the response is performed by the vehicle; selecting, based on the predicted future states and predicted first zone future safety values, one of the alternative responses for the vehicle; and providing the selected alternative response to a drive control system of the vehicle to cause the drive control system of the vehicle to correspondingly adjust the spacing between the vehicle and the moving object.

2 . The method of claim 1 further comprising, predicting, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in the set of multiple alternative responses, a second zone future safety value corresponding to a second safety zone of the vehicle, the second safety zone of the vehicle being distinct from the first safety zone of the vehicle, the second zone future safety value indicating a safety level of the vehicle for the second safety zone if the response is performed by the vehicle, wherein the selected alternative response is selected also based on the predicted second zone future safety values.

3 . The method of claim 2 wherein the selected alternative response is also based on a target state for the vehicle, and the method comprises controlling the vehicle to perform the selected alternative response for the vehicle.

4 . The method of claim 3 wherein
the first safety zone is located in front of the vehicle and the predicted first zone future safety value for each response in the set of multiple alternative responses indicates a likelihood of a leading vehicle being present in the first safety zone, and
the second safety zone is located behind the vehicle and the predicted second zone future safety value for each response in the set of multiple alternative responses indicates a likelihood of a trailing vehicle being present in the second safety zone.

5 . The method of claim 4 wherein the future state, the first zone future safety value and the second zone future safety value for each response in the set of multiple alternative responses are predicted using one or more trained neural networks.

6 . The method of claim 5 wherein the current vehicle state includes: (i) a speed of the vehicle; (ii) a distance from the vehicle to any leading vehicle detected in front of the vehicle; and (iii) a distance from the vehicle to any trailing vehicle detected in back of the vehicle.

7 . The method of claim 6 wherein the current vehicle state includes a current first zone safety value indicating if any leading vehicle is currently present in the first safety zone and a current second zone safety value indicating if any trailing vehicle is currently present in the second safety zone.

8 . The method of claim 1 wherein selecting the selected alternative response for the vehicle comprises: selecting one of the multiple alternative responses for which the predicted future state satisfies a state condition and the predicted first zone future safety value satisfies a first zone safety condition.

9 . The method of claim 8 wherein selecting the selected alternative response for the vehicle is performed by a fuzzy inference system, the selecting further comprising:
receiving the predicted future states and first zone future safety values, wherein for each response, the predicted future state includes a vehicle speed prediction, and the predicted first zone future safety value includes a vehicle safety level;
performing fuzzification of the vehicle speed predictions to map the vehicle speed predictions to target speed truth values that denote closeness of the vehicle speed predications to a target speed;
performing fuzzification of the predicted first zone future safety values to map the first zone future safety values to safety fuzzy truth values;
based on the target speed truth values and the safety fuzzy truth values, performing fuzzy inference to generate a goal fuzzy set;
defuzzifying the goal fuzzy set to select, as the selected alternative response for the vehicle, a best response from the set of multiple alternative responses to satisfy the state condition and the first zone safety condition.

10 . The method of claim 1 , wherein the first zone future safety value for the vehicle is predicted by a safety state function of the machine learning model.

11 . An adaptive spacing predictive control system for controlling a vehicle to adaptively control spacing between the vehicle and a moving object in an operating environment of the vehicle, comprising:
a processor system; a memory coupled to the processor system, the memory tangibly storing thereon executable instructions that, when executed by the processor system, cause the processor system to:
train a machine learning model using reinforcement learning to select vehicle responses based on predicted vehicle states and safety values, wherein the machine learning model is trained using a general value function (GVF) framework;
determine a current vehicle state of the vehicle and a current object state of the moving object based on sensor data captured by sensors of the vehicle;
predict, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in a set of multiple alternative responses, a future state for the vehicle corresponding to the response;
predict, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in the set of multiple alternative responses, a first zone future safety value corresponding to a first safety zone of the vehicle, the first zone future safety value indicating a safety level of the vehicle for the first safety zone if the response is performed by the vehicle; select, based on the predicted future states and predicted first zone future safety values, one of the alternative responses for the vehicle; and provide the selected alternative response to a drive control system of the vehicle to cause the drive control system of the vehicle to correspondingly adjust the spacing between the vehicle and the moving object.

12 . The system of claim 11 wherein the executable instructions, when executed by the processor system, also cause the processor system to:
predict, by learned predictive functions of the machine learning model, based on the current vehicle state and current object state, for each response in

Assignees

• Graves Daniel Mark
• Rezaee Kasra
• Huawei Tech Co Ltd

Inventors

• Graves Daniel Mark
• Rezaee Kasra

Classifications

• B60W2754/30

  Longitudinal distance · CPC title

• B60W2554/802

  Longitudinal distance · CPC title

• B60W30/16 Primary

  Control of distance between vehicles, e.g. keeping a distance to preceding vehicle · CPC title

• B60W2754/10

  Spatial relation or speed relative to objects · CPC title

• B60W2554/80

  Spatial relation or speed relative to objects · CPC title