Use Case #4:
Automated Sensor KPI Calculation
![](https://i0.wp.com/ottometric.com/wp-content/uploads/2023/09/SEN_EDIT_r2.png?resize=150%2C150&ssl=1)
Problem:
The conventional approach for sensor KPI calculation is to capture thousands of hours of road test data and send it out to large teams of annotators who draw bounding boxes around objects and estimate distances and speeds, creating a “ground truth” against which sensor performance can be measured. This is a costly and error prone process, which requires a huge amount of manual effort to create the annotated images and review and fix annotation errors. Once the ground truth has been vetted, validation teams typically create scores of custom scripts to run against the ground truth to measure the accuracy of the sensor perception in a variety of conditions. This is also a highly manual-intensive task with limited reuse between programs.
Solution:
The Ottometric platform automatically computes a synthetic ground truth by fusing LiDAR, camera and other sensor data, and automatically calculates the KPIs for the perception system. It detects true and false positives as well as true and false negatives. KPIs are summarized in a comprehensive report, and failures are indicated in a timeline display and can be examined in detail using the Ottoviz data visualizer.
Benefits:
• Reduces the cost of sensor validation by 75% compared to traditional manual annotation and KPI analysis
• Shortens time to market by automating data analysis of huge data sets