URMAP® 360
A single radar sees only a slice of the world. URMAP 360 fuses the object tracks from a whole suite of radars into one shared 360° scene — keeping every object’s identity as it passes from one radar to the next, with no blind spots and no drop-outs at the hand-off.
360°
surround coverage
0
blind spots
0-drop
object persistence at hand-off
On-radar
decentralised fusion
Applications
Where multiple radars need to see one shared scene.
Automotive
Autonomy platforms with multi-radar architectures — front, side, rear and corner sensors. URMAP 360 turns the array into one perception system instead of N independent ones, so the platform sees a single coherent scene.
Industrial
Mining haul-trucks, agricultural machinery, site-safety systems on heavy plant. Multi-radar suites give full-perimeter situational awareness around large vehicles, with workers and other equipment tracked continuously from front to side to rear.
Smart city
Roadside and infrastructure networks — radars mounted along a highway or on either side of an overpass. URMAP 360 hands a vehicle from one radar to the next, so it is followed continuously as it approaches and moves away from the bridge — no central server, just radars cooperating.
How it works
Multi-radar fusion that keeps every object as one object.
STEP 01
Put every radar on one map
Each radar sees the scene from its own position, at its own moment in time. Before their detections can be combined, URMAP 360 places every radar on one shared map and one shared clock — so a vehicle picked up by one radar lines up exactly with the same vehicle its neighbour sees. The radars sort this out between themselves, with no central computer to set up.
STEP 02
Fuse objects, not raw data
Each radar already does its own detecting and tracking. URMAP 360 combines those finished object tracks — not the raw radar signal — matching the same object across radars and merging it where their views overlap, so nothing is counted twice. Because it works with objects, it runs right on the radars themselves, with no central fusion box.
STEP 03
Hand off without dropping a track
As a vehicle leaves one radar’s view and enters the next — often with a short gap in between — URMAP 360 keeps the same identity on it. It predicts where the vehicle should reappear, matches it to the one the next radar picks up, and double-checks using the object’s type. The track carries straight through with the same ID, speed and history — no flicker, no lag, no double-counting.
Featured deployment
Unified 360° situational awareness.
URMAP 360 is the intelligence layer that turns a collection of independent radars into one cohesive perception system. It runs decentralised — on each radar’s onboard compute — fusing object tracks across the array into a single world-frame scene with persistent identities, with no central fusion computer required.
- Zero-drop object persistence across the full 360° perimeter
- Decentralised — runs on each radar’s onboard compute
- Cross-platform: automotive, industrial and smart-city infrastructure
Integration
Runs on-radar, peer-to-peer
URMAP 360 runs on each radar’s onboard compute alongside the single-radar tracker — decentralised, with no central fusion node; radars cooperate peer-to-peer over DDS. It shares URMAP’s universal training base and the same downstream interface, so adding it to an existing URMAP deployment needs no architectural change.
- Runs on-radar
- Multi-radar PTP synchronisation
- DDS peer discovery & world-frame track output
- Shared training base with URMAP
- Same downstream interface as URMAP
Questions, answered
URMAP 360 FAQ
URMAP 360 is a multi-radar surround-perception layer; URMAP is the single-radar perception model. Each radar runs URMAP locally, and URMAP 360 fuses their object tracks across the array — front, side, rear, corner — into one shared, world-frame scene with a single persistent identity per object as it moves between sensors. Same universal training base, different problem shape.
On the radar — by design. URMAP 360 runs decentralised: a fusion service on each radar’s onboard compute, next to that radar’s tracker. Each radar publishes its tracks in a shared world frame and subscribes to its neighbours’, so the array fuses peer-to-peer with no central server. A single radar still works standalone, and more radars can join automatically.
At the object level. (1) Where two radars’ fields of view overlap, the tracks they each report for the same object are associated and merged into one, rather than double-counted. (2) Across the gap between fields of view, proprietary hand-off logic keeps an object on the same track ID, same velocity estimate and same trajectory history as it moves from one radar to the next — zero-drop persistence across the perimeter.
One scene from many radars.
If you’re building a multi-radar platform — on a vehicle, on heavy plant, or across infrastructure — URMAP 360 is the model that turns the sensor array into a single coherent perception system.