This n8n workflow automates the process of annotating, storing, and querying Advanced Driver-Assistance System (ADAS) event data, making it highly useful for automotive data analysis and incident management. The workflow begins with a webhook trigger that receives ADAS event data via an HTTP POST request. This data is then subdivided into smaller chunks using a text splitter node to facilitate detailed analysis.
The smaller text segments are transformed into embeddings using Cohere’s API, which enables semantic understanding of the data. These embeddings are stored in a Supabase vector database, supporting efficient similarity searches later. When a new query or incident occurs, the system queries the vector database to find relevant similar data points.
A language model (OpenAI) powered agent then processes this information, aided by a memory buffer to maintain context across interactions. The agent is orchestrated through a custom prompt, ensuring precise responses or annotations for the ADAS events.
Finally, the workflow logs pertinent information or results into a Google Sheet for record-keeping or further analysis. The entire process facilitates real-time, intelligent annotation and querying of ADAS event data for automotive safety and research applications.
This workflow is particularly suitable for automotive companies or researchers aiming to automate the analysis and documentation of vehicle sensor alerts or incident data, providing faster insights and improved data management.
Reviews
There are no reviews yet.