Autonomous Vehicles

ASOF JOIN for multi-sensor time sync. Parquet HDB for petabyte-scale driving log replay.

---

Why AV Teams Choose ZeptoDB

Sensor Time Sync

ASOF JOIN aligns LiDAR, camera, radar, and IMU streams with nanosecond precision.

Driving Log Replay

Query historical driving episodes from Parquet on S3. Replay at original speed or accelerated.

Zero-Copy ML

522ns from query result to PyTorch tensor for perception model training.

Fleet Scale

Multi-node cluster handles telemetry from thousands of vehicles simultaneously.

---

Use Cases

Multi-Sensor Fusion

• Time-align LiDAR point clouds, camera frames, radar returns, and IMU data
• Window JOIN with ±N ms tolerance for sensor clock drift compensation
• Real-time fusion pipeline for online perception

Scenario Mining

• SQL queries over petabytes of driving logs stored as Parquet on S3
• Find specific scenarios: “all left turns at intersections with pedestrians in rain”
• ASOF JOIN to correlate vehicle state with map/weather data

Training Data Pipeline

• Query → NumPy → PyTorch in microseconds via zero-copy
• Batch extraction via Arrow Flight for distributed training
• Reproducible dataset versioning with Parquet snapshots

---

Example: Multi-Sensor Alignment

-- Align all sensor streams to LiDAR timestamps
SELECT
    l.ts AS lidar_ts,
    l.points,
    c.frame AS camera_frame,
    r.detections AS radar_detections,
    i.accel_x, i.accel_y, i.accel_z,
    i.yaw_rate,
    g.lat, g.lon, g.heading
FROM lidar_scans l
ASOF JOIN camera_frames c ON l.vehicle_id = c.vehicle_id AND l.ts >= c.ts
ASOF JOIN radar_returns r ON l.vehicle_id = r.vehicle_id AND l.ts >= r.ts
ASOF JOIN imu_readings i ON l.vehicle_id = i.vehicle_id AND l.ts >= i.ts
ASOF JOIN gps_fixes g ON l.vehicle_id = g.vehicle_id AND l.ts >= g.ts
WHERE l.vehicle_id = 'vehicle-042'
  AND l.ts BETWEEN '2026-03-15 14:00:00' AND '2026-03-15 14:30:00'

---

Architecture

Vehicle Edge Node              Cloud Cluster              Training
─────────────────    ─────────────────────────    ─────────────────
LiDAR      ──→  │                           │
Camera     ──→  │ ZeptoDB Edge              │
Radar      ──→  │   ↓ WAL + Parquet        │──→  S3 Parquet HDB
IMU        ──→  │   ↓ Sync to cloud        │──→  ZeptoDB Cluster
GPS        ──→  │                           │──→  Arrow Flight
                 ─────────────────────────         ↓
                                               PyTorch Training