"* Preconditions: Radar detection + tracking algorithm (including ML based postprocessing/classification) shall achieve 99% vehicle counting accuracy and speed measurement error within 2kph/2% (bigger of) under various traffic scenarios where vehicles appear very close to each other and going with approximately equal speed (side by side). * Input conditions / steps: radar data from simulator/real-world with significant occurrence of above mentioned cases. * Expected result: The system shall meet counting and speed measurement accuracy when close passing vehicles (side by side) are present. "