Arbe’s ultra-HD radar perception generates dense, long-range point-cloud data with direct range measurement per cell, forming a continuously updated occupancy grid that classifies space as occupied, free, or unknown. High spatial resolution defines the boundaries of occupied space, while fine Doppler resolution distinguishes stationary from dynamic objects. Together, they enable reliable free space mapping in any lighting or weather condition, supporting certified Eyes-Off operation.

Arbe resolves low radar signature objects, elevated hazards, lost cargo, and road construction through high spatial resolution and dense detections per object. Fine Doppler sensitivity enables accurate identification of slow-moving or stationary vulnerable road users, supporting safe automated operation in complex real-world conditions.

Arbe’s ultra-HD radar for perception delivers high spatial resolution in all dimensions, long-range reliability, and consistent performance regardless of lighting or weather conditions. Its high Doppler resolution further enables efficient separation of stationary and dynamic objects. Together, these capabilities make imaging radar a mandatory sensor for real-time free space mapping in certified Eyes-Off systems.

Arbe’s ultra-HD radar contributes direct measurements of distance, velocity, and elevation to the sensor fusion stack. This independent data source stabilizes perception in low visibility and supports the redundancy required for Eyes-Off, Level 3 driving systems.

Through direct velocity measurement and high-resolution spatial sampling, Arbe’s radar enables continuous multi-object tracking across frames. This allows automated systems to track vehicles and obstacles reliably at highway speeds, even in dense or rapidly changing traffic.

Ultra-high angular resolution and hundreds of detections per object allow Arbe’s radar to distinguish adjacent vehicles, barriers, and stationary hazards that conventional radar systems merge together. This precise object separation improves scene understanding in dense highway environments and helps prevent incorrect automated responses.