View source for 20240028601. METHODS AND APPARATUS FOR NATURAL LANGUAGE-BASED SAFETY CASE DISCOVERY TO TRAIN A MACHINE LEARNING MODEL FOR A DRIVING SYSTEM simplified abstract (PlusAI, Inc.)

=METHODS AND APPARATUS FOR NATURAL LANGUAGE-BASED SAFETY CASE DISCOVERY TO TRAIN A MACHINE LEARNING MODEL FOR A DRIVING SYSTEM=


==Organization Name==

[[:Category:PlusAI, Inc.|PlusAI, Inc.]]


[[Category:PlusAI, Inc.]]

==Inventor(s)==

[[:Category:Hao Zheng of Los Altos Hills CA (US)|Hao Zheng of Los Altos Hills CA (US)]][[Category:Hao Zheng of Los Altos Hills CA (US)]]

[[:Category:Yang Yang of Fremont CA (US)|Yang Yang of Fremont CA (US)]][[Category:Yang Yang of Fremont CA (US)]]

==METHODS AND APPARATUS FOR NATURAL LANGUAGE-BASED SAFETY CASE DISCOVERY TO TRAIN A MACHINE LEARNING MODEL FOR A DRIVING SYSTEM - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240028601 titled 'METHODS AND APPARATUS FOR NATURAL LANGUAGE-BASED SAFETY CASE DISCOVERY TO TRAIN A MACHINE LEARNING MODEL FOR A DRIVING SYSTEM

==Simplified Explanation==

The patent application describes a safety case discovery system that utilizes a scenario framework and safety protocols to identify edge cases in vehicle operation. The system collects sensor data from at least one sensor during vehicle operation and stores it in a data warehouse. The data warehouse can be queried based on predefined scenario descriptions to retrieve a subset of records that are relevant to the scenario. The system then ranks these records based on their relevancy and removes any duplicates to identify edge cases. The safety protocol is updated based on these edge cases. Additionally, the system can train a machine learning model for optimizing the driving system using the identified edge cases.

* The system collects sensor data during vehicle operation and stores it in a data warehouse.
* The data warehouse can be queried based on predefined scenario descriptions.
* The system ranks the retrieved records based on their relevancy to the scenario.
* Duplicate records are removed to identify edge cases.
* The safety protocol is updated based on the identified edge cases.
* The system can train a machine learning model using the edge cases to optimize the driving system.

== Potential Applications ==
* Enhancing vehicle safety by identifying and addressing edge cases in driving scenarios.
* Improving the performance and efficiency of autonomous driving systems through machine learning.

== Problems Solved ==
* Identifying edge cases in vehicle operation that may pose safety risks or require special handling.
* Streamlining the process of discovering and addressing safety issues in driving scenarios.

== Benefits ==
* Increased safety for vehicles by addressing potential edge cases.
* Improved performance and efficiency of autonomous driving systems through machine learning optimization.


==Original Abstract Submitted==

a safety case discovery system includes a scenario framework and safety protocols for edge cases. the safety case discovery system receives sensor data generated by at least one sensor during operation of a vehicle and stores the sensor data in a data warehouse. the data warehouse can be queried based on a predefined scenario description to produce a subset of records which are ranked based on a relevancy of the records to the predefined scenario description. the safety case discovery system deduplicates the ranked results to produce edge cases and updates the safety protocol. the safety case discovery system can train a machine learning model vehicle based on the edge cases, to produce a trained machine learning model for optimizing the driving system.



[[Category:G06F16/2457]]
[[Category:B60W60/00]]
[[Category:G06F16/242]]