Finding
Radar is the sensor that works in any weather, and 4D imaging radar is closing the detail gap with lidar
Automotive radar measures distance and relative speed directly using radio waves in the 77 GHz band, and it keeps working at night, in glare, rain, fog, spray and dust. That makes it the default partner for cameras in driver assistance: automatic emergency braking, adaptive cruise and blind-spot monitoring all rely on it. Its weakness has been low angular resolution, which makes a stopped car hard to tell from an overhead sign.
4D imaging radar adds elevation and many more virtual channels, producing a sparse point cloud closer to lidar's at a fraction of the cost. Chinese Level 2+ systems increasingly pair cameras with 4D radar, and robotaxi stacks use it alongside lidar for redundancy in bad weather. Suppliers include Continental, Bosch, Aptiv, ZF and Arbe in the West, and Huawei and a growing group of domestic makers in China. Radar chips are built on RF CMOS and silicon-germanium processes, which ties radar supply to a different part of the semiconductor chain than cameras or compute.
Types
Radar types
| Type | Coverage | Uses | Limit |
|---|---|---|---|
| Short-range | Corners and sides | Blind-spot monitoring, rear cross-traffic, parking | Short reach |
| Medium-range | Sides and surround | Lane-change support, surround awareness | Less range than long-range |
| Long-range | Forward path | Adaptive cruise, emergency braking, highway | Narrow field; poor elevation |
| 4D imaging | Forward and surround | Detailed scene in all weather; L2+ and L4 | Higher cost and compute |
Sources
Sources and scope
- Scope
- Automotive radar for driver assistance and autonomy
- Sources
- Supplier specifications; ElectronsX analysis
- Updated
- October 2026