Verified Inference Engine with Theorem-Proved Input-Output Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle inference engines face challenges in ensuring high assurance and reliability, particularly in meeting stringent standards like DO-178C Level A and EAL-7, due to the risk of executing contradictory commands and making decisions based on flawed data.
Innovation Solution
Implementing a verified inference engine with input and output validators, and a rule validator, which uses automated theorem proving to ensure logical consistency and mathematical correctness, thereby validating inputs and outputs to prevent catastrophic failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated theorem proving is used to verify inference engine correctness, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by performing automated theorem proving during the development and verification phase before deployment. The inference engine is formally verified against its specification using automated theorem proving tools, ensuring correctness is established beforehand rather than through runtime checks or post-deployment testing.
Solution Approach 2:
The patent uses an intermediary approach by introducing a formal specification as a mediator between the inference engine implementation and the desired correctness properties. The automated theorem proving tool acts as an intermediary that formally verifies whether the implementation satisfies the specification, providing mathematical guarantees without requiring direct manual verification.
2Reliability
If input validation is implemented to ensure reliable data, then reliability is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing input validation before the inference engine processes the data. The system validates that inputs conform to expected formats and ranges prior to inference execution, preventing invalid data from entering the processing pipeline and avoiding costly reprocessing or error handling later.
3Reliability
If output validation is performed to prevent forbidden maneuvers, then safety is improved, but system response time decreases
Solution Approach 1:
The patent applies preliminary action by establishing output validation rules and constraints before the inference engine generates outputs. The system pre-defines forbidden maneuvers and safe operating boundaries, allowing rapid validation of outputs against these pre-established criteria rather than performing complex safety analysis after each inference.
4Reliability
If rule consistency validation is performed using automated theorem proving, then reliability is improved, but computational resources increase
Solution Approach 1:
The patent applies preliminary action by performing rule consistency validation during the rule base construction and verification phase. The system uses automated theorem proving to verify that rules are logically consistent with the specification before deploying the inference engine, eliminating the need for continuous runtime consistency checks and reducing ongoing computational energy consumption.
Data Source
AI summary
An autonomous vehicle system includes one or more output devices, one or more input sources, and a controller. In some embodiments, a verified inference engine is used by the controller to generate outputs for the output devices from inputs received from the input sources. The inference engine may be verified to be mathematically correct with an automated theorem proving tool. The automated theorem proving tool may verify that the inference engine meets the design requirements of standards such as DO-178C Level A and/or EAL-7. The controller is configured to validate the inputs received from the input sources, store the validated inputs in a fact base, generate outputs from the validated inputs, validate the one or more outputs, and provide the one or more validated outputs to the output devices.