Vehicle decision making using sequential information probing

Patent metadata

Field	Value
Publication number	US-12559113-B2
Application number	US-202418429196-A
Country	US
Kind code	B2
Filing date	Jan 31, 2024
Priority date	Jan 31, 2024
Publication date	Feb 24, 2026
Grant date	Feb 24, 2026

Abstract

Vehicle decision-making is analyzed and can be used to modify a decision-making process. For subsets of features comprising a vehicle operational scenario, a first value is generated that quantifies behavior of an artificial intelligence (AI) agent as the AI agent performs a sequence of actions within a first world model based on a complete set of observations for the subset of features. A first world model is a copy of a second world model for sequential decision making. A second value is generated that quantifies behavior of an AI agent as the AI agent performs a sequence of actions in the second world model based on an incomplete set of observations for the subset of features. A difference between the first and second values determines the impact of individual features on the AI agent within the second world model. A decision-making process of the AI agent can be updated.

First claim

What is claimed is:

1 . A method, comprising:
determining a first world model, wherein the first world model is a copy of a second world model for sequential decision making with incomplete state information; for respective subsets of features comprising a vehicle operational scenario within a vehicle transportation system:
generating a first value quantifying behavior of a first artificial intelligence (AI) agent as the first AI agent performs a sequence of actions responsive to the vehicle operational scenario using the first world model, wherein the first AI agent includes a complete set of observations for the subset of features within the first world model;
generating a second value quantifying behavior of a second AI agent as the second AI agent performs a sequence of actions responsive to the vehicle operational scenario using the second world model, wherein the second AI agent includes an incomplete set of observations for the subset of features within the second world model; and
determining a difference between the first value and the second value;
calculating an impact of individual features on the second AI agent within the second world model using the differences; and selectively, based on the impact of individual features, updating a decision-making process of the second AI agent.

2 . The method of claim 1 , wherein the sequence of actions performed by the first AI agent includes a pre-defined number of actions.

3 . The method of claim 1 , wherein generating the second value comprises:
introducing the second AI agent into the second world model; and allowing the second AI agent to perform the sequence of actions within the second world model.

4 . The method of claim 1 , wherein generating the first value comprises:
introducing the first AI agent into the first world model; allowing the first AI agent to perform the sequence of actions within the first world model; and updating a belief state of the second AI agent using the first AI agent.

5 . The method of claim 4 , wherein generating the first value further comprises:
providing the first AI agent with information that the first world model is a copy of the second world model.

6 . The method of claim 4 , wherein updating the belief state of the second AI agent using the first AI agent may be performed after each action in the sequence of actions.

7 . The method of claim 1 , wherein the second AI agent is used, by a control system of the vehicle, to control the vehicle.

8 . An apparatus, comprising:
a memory; and a processor configured to execute instructions stored in the memory to: determine a first world model, wherein the first world model is a copy of a second world model for sequential decision making with incomplete state information; for respective subsets of features comprising a vehicle operational scenario within a vehicle transportation system:
generate a first value quantifying behavior of a first artificial intelligence (AI) agent as the first AI agent performs a sequence of actions responsive to the vehicle operational scenario using the first world model, wherein the first AI agent includes a complete set of observations for the subset of features within the first world model;
generate a second value quantifying behavior of a second AI agent as the second AI agent performs a sequence of actions responsive to the vehicle operational scenario using the second world model, wherein the second AI agent includes an incomplete set of observations for the subset of features within the second world model; and
determine a difference between the first value and the second value;
calculate an impact of individual features on the second AI agent within the second world model using the differences; and selectively, based on the impact of individual features, update a decision-making process of the second AI agent.

9 . The apparatus of claim 8 , wherein the sequence of actions performed by the first AI agent includes a pre-defined number of actions.

10 . The apparatus of claim 8 , wherein to generate the second value includes instructions to:
introduce the second AI agent into the second world model; and allow the second AI agent to perform the sequence of actions within the second world model.

11 . The apparatus of claim 8 , wherein to generate the first value includes instructions to:
introduce the first AI agent into the first world model; allow the first AI agent to perform the sequence of actions within the first world model; and update a belief state of the second AI agent using the first AI agent.

12 . The apparatus of claim 11 , wherein to generate the first value includes instructions to:
provide the first AI agent with information that the first world model is a copy of the second world model.

13 . The apparatus of claim 11 , wherein to update the belief state of the second AI agent using the first AI agent may be performed after each action in the sequence of actions.

14 . The apparatus of claim 8 , wherein the second AI agent is used, by a control system of the vehicle, to control the vehicle.

15 . A non-transitory computer-readable medium storing instructions operable to cause one or more processors to perform operations comprising:
determining a first world model, wherein the first world model is a copy of a second world model for sequential decision making with incomplete state information; for respective subsets of features comprising a vehicle operational scenario within a vehicle transportation system:
generating a first value quantifying behavior of a first artificial intelligence (AI) agent as the first AI agent performs a sequence of actions responsive to the vehicle operational scenario using the first world model, wherein the first AI agent includes a complete set of observations for the subset of features within the first world model;
generating a second value quantifying behavior of a second AI agent as the second AI agent performs a sequence of actions responsive to the vehicle operational scenario using the second world model, wherein the second AI agent includes an incomplete set of observations for the subset of features within second world model; and
determining a difference between the first value and the second value;
calculating an impact of individual features on the second AI agent within the second world model using the differences; and selectively, based on the impact of individual features, updating a decision-making process of the second AI agent.

16 . The non-transitory computer-readable medium of claim 15 , wherein the sequence of actions performed by the first AI agent includes a pre-defined number of actions.

17 . The non-transitory computer-readable medium of claim 15 , wherein generating the second value comprises:
introducing the second AI agent into the second world model; and allowing the second AI agent to perform the sequence of actions within the second world model.

18 . The non-transitory computer-readable medium of claim 15 , wherein generating the first value comprises:
introducing the first AI agent into the first world model; allowing the first AI agent to perform the sequence of actions within the first world model; and updating a belief state of the second AI agent using the first AI agent.

19 . The non-transitory computer-readable medium of claim 18 , wherein generating the first value further comprises:
providing the first AI agent with information that the first world model is a copy of the second world model.

Assignees

• Nissan North America Inc
• Univ Massachusetts

Inventors

• Vazquez-Chanlatte Marcell Jose
• Witwicki Stefan
• Zilberstein Shlomo
• Mahmud Saaduddin

Classifications

• G05B13/0265

  the criterion being a learning criterion · CPC title

• B60W2050/0028

  Mathematical models, e.g. for simulation · CPC title

• B60W60/00

  Drive control systems specially adapted for autonomous road vehicles · CPC title

• B60W60/001

  Planning or execution of driving tasks · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title