Causal AI Policy Explanations for Autonomous Vehicle Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional AI systems used in autonomous vehicles lack the ability to provide clear explanations for their decision-making processes, which can lead to a lack of trust from users.
Innovation Solution
A method is developed to generate a causal machine learning model that explains AI policies of behavior by identifying state factors and their responsibility scores, allowing for real-time causal explanations of actions taken by autonomous vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional AI systems are used for autonomous vehicle decision-making, then automated driving functionality is achieved, but the ability to provide clear explanations for decisions is lost
Solution Approach 1:
The patent introduces an intermediary explanation generation module that sits between the automated AI decision-making system and the user. This module translates complex AI decisions into human-understandable explanations by identifying critical state factors and generating natural language descriptions, thus preserving information without compromising automation.
Solution Approach 2:
The explanation system segments the complex AI decision-making process into identifiable state factors and their contributions. By breaking down the decision into discrete elements (state factors, their values, and explanatory text), the system makes automated decisions transparent while maintaining full automation capability.
2Measurement precision
If complex AI algorithms are used to determine driving behaviors, then decision accuracy is improved, but the complexity of the system increases
Solution Approach 1:
The patent extracts the explanation generation function as a separate module from the complex AI decision-making system. This extraction allows the main AI algorithm to maintain its complexity for accurate decision-making while the separate explanation module handles the complexity of translating decisions into understandable formats, effectively managing system complexity.
3Reliability
If real-time causal explanations are provided for vehicle actions, then user trust is enhanced, but computational resources are consumed
Solution Approach 1:
The explanation system applies partial action by focusing only on the most critical state factors that contribute to the decision, rather than explaining every aspect of the complex AI processing. By identifying and explaining only the top contributing factors (e.g., 3-5 key state factors), the system provides sufficient explanation to build trust while consuming minimal computational energy.
Data Source
AI summary
A method includes generating a data structure including states and actions to be executed at those states as determined by an AI policy of behavior, determining, with a first computing system that is offline, state factors associated with the states and responsibility scores for the state factors, each responsibility score indicating a causal impact for each of the actions associated with one of the states, generating, with the first computing system, a causal ML model based on the state factors and the responsibility scores, determining, with a second computing system that is online based on the causal ML model, state factors associated with a current state, and identifying one or more of the state factors as a causal reason for an action resulting from the current state. Other example methods and systems for providing explanation of an AI policy of behavior with causal reasoning are also disclosed.