OBJECT_TRACKING_MULTI_SENSOR_FUSION

Sensor Fusion for Target Tracking

0. Introduction :-

We have an array of sensors installed around the ego vehicle for ADAS/AD applications. The goal of this project is to estimate Valid Object Tracks by fusing the measurements from the sensors.

Fig 1 : (Inputs) Measurements : Radar Reflection, Camera Detections, Lane Line Clothoid Parameters

1. Estimation Output :-

1.1 Object Tracks :-

1.2 Object Tracks + Object Trajectories:-

1.3 Object Tracks + Object Trajectories + Road Model:-

1.4 Object Tracks + Object Trajectories + Road Model + Mapping:-

2. Fusion Architecture Overview :-

3. High Level Fusion Architecture:-

4. Module Level Fusion Architecture:-

4.1 Preprocessing Subsystem:-

4.2 Radar Fusion Subsystem :-

4.3 Camera Fusion Subsystem :-

4.4 Radar & Camera Fusion Subsystem :-