Benchmark RadarSIREN-Bench-v1
SIREN-Bench-v1 is a benchmark for emergency-vehicle (EMV) interactions, built on a SUMO-CARLA co-simulation platform. It includes seven parameterized interaction templates across emergency levels L1-L3 and three behavior families, with synchronized sensor observations and simulator-native annotations. Supports tasks such as 3D object detection, trajectory prediction, and vision-language risk understanding.
- Released
- 2026-08-25
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Evaluating EMV interactions requires behavior-level control over both EMV privileges and civilian responses, which existing benchmarks lack. SIREN-Bench-v1 provides a reproducible co-simulation environment with behavior-dependent failure modes, enabling standard evaluation for safety-critical autonomous driving research.
Motivation
Emergency vehicles (EMVs) can reorganize surrounding traffic as civilian vehicles brake, change lanes, or form rescue corridors in response to their passage.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.