XTRACTIS FOR adas & autonomous vehicle

Emergency Detection for an Automatic Braking Assist

Benchmark vs. Random Forests, Boosted Trees & Neural Networks

How to automatically, efficiently and transparently diagnose driving situations to activate the emergency braking, only from the car recordings without camera, radar, or lidar?

Goals & benefits

Identify the car parameters involved in the driving situation diagnosis and enhance technical knowledge by helping engineers understand the causal relationships between specific parameters, their combination, and the occurrence of an emergency.

Approach the perfect model: the slightest mistake can be fatal. Help engineers design reliable intelligible autonomous vehicles that assist the driver efficiently according to their driving style. Intelligible means that the internal decision logic of the decision system is explicit.

Enforce the use of stable and transparent models audited by the domain expert and certified by the regulator before embedding them in the vehicle.

Challenge xtractis to find better models than those we crafted “by hand” in 1999 and far quicker!


We get a Predictive Model that is:


A Decision System composed of 25 unchained gradual rules, each rule uses some of the 12 variables that xtractis identified as significant.


Very good Real Performance on the Test dataset.

Efficient & Operational.

Running in real-time up to 70,000 predictions per second (i7 @2.5GHz with 8 physical cores), offline or online (API).

Use Case 11/2021 (v1.3)

Results by
XTRACTIS® GENERATE 11.3.40047 (11/2021)


  1. Problem Definition
  2. Xtractis Solution
  3. Top-Model Induction
  4. Explained Predictions for 3 Cases
  5. Top-Models Benchmark
  6. Analysis of the Models’ Temporal Response