MVP Roadmap: Radar-Only Multi-static Tracking

[jonnyspicer]

MVP Vision

Create a minimum viable passive radar tracking system that can:

• Track aircraft using ONLY radar detections (no ADS-B dependency)
• Process multi-static radar measurements properly
• Display tracks on a 3D map interface
• Provide basic track lifecycle management

Current System Cannot Operate Without ADS-B

The system has a fundamental dependency on ADS-B for target association. Without ADS-B truth data, no targets are tracked.

Current Status ✅

• Stone Soup integration completed (Replace custom tracker with Stone Soup MultiTargetTracker #1) - Modern tracking foundation
• 3D visualization with Cesium map display working
• Localization algorithms implemented and working
• Service architecture with API, event, and visualization components
• ADS-B + radar tracking working when ADS-B is available

🔴 MVP Gaps

#20 - RadarOnlyAssociator 🚨 CRITICAL

• Problem: Only AdsbAssociator exists, requires ADS-B truth data
• Impact: System cannot track targets using only radar detections
• Solution: Implement associator that groups radar detections without prior target knowledge
• Priority: Must complete FIRST - everything else depends on this

#21 - TDOA/FDOA Processing 🟡 HIGH PRIORITY

• Problem: Multi-static radar needs relative measurements between radar pairs
• Impact: Suboptimal positioning from individual delay/Doppler measurements
• Solution: Implement Time/Frequency Difference of Arrival processing
• Priority: Required for proper multi-static radar operation

#18 - Puppeteer Testing 🧪 DEVELOPMENT EFFICIENCY

• Purpose: Enable faster development with automated testing
• Benefit: Test-driven development for radar-only functionality
• Priority: Complete early to support other development

📉 Downgraded Issues (Code Quality, Not Functionality)

These were incorrectly identified as MVP-critical. They're valuable post-MVP enhancements:

• Assess and improve localization accuracy (post-radar-association) #9 - Localization accuracy assessment → Post-MVP (current geometric methods may be adequate)
• Add statistical gating optimization (post-MVP enhancement) #10 - Statistical gating → Post-MVP (optimization, not core functionality)
• Refactor track lifecycle for better code organization (post-MVP) #11 - Track lifecycle refactor → Post-MVP (code organization, current system works)
• Optimize association algorithm with better assignment (post-MVP) #12 - Association algorithm optimization → Post-MVP (need basic radar association first)

Corrected Implementation Order

Phase 1: Enable Radar-Only Operation (4-6 weeks)

1. Set up Puppeteer MCP server for automated testing of 3lips package #18 - Testing infrastructure (1-2 weeks)
2. Implement RadarOnlyAssociator for pure radar tracking #20 - RadarOnlyAssociator (2-3 weeks) - CRITICAL
3. Implement TDOA/FDOA processing for multi-static radar measurements #21 - TDOA/FDOA processing (1-2 weeks) - HIGH

Phase 2: Post-MVP Enhancements (Future)

4. Assess localization accuracy with real radar-only data (Assess and improve localization accuracy (post-radar-association) #9)
5. Optimize association and gating (Add statistical gating optimization (post-MVP enhancement) #10, Optimize association algorithm with better assignment (post-MVP) #12)
6. Improve code organization (Refactor track lifecycle for better code organization (post-MVP) #11)

Success Criteria for MVP

• System can track targets using only radar detections (no ADS-B required)
• Works with synthetic-adsb test data in radar-only mode
• Produces reasonable tracks from multi-static radar measurements
• Real-time performance maintained
• 3D visualization displays radar-only tracks

Key Insight: Functionality vs Optimization

MVP = Working radar-only tracking capability

• Get radar-only association working (functionality gap)
• Use existing localization and tracking (already works)
• Optimize later (statistical gating, better algorithms, etc.)

Current Reality Check ✅

The system already has:

• Localization algorithms
• Working Stone Soup tracking
• Good visualization
• Solid architecture

Missing piece: Ability to operate without ADS-B truth data

Getting Started

1. Set up Puppeteer MCP server for automated testing of 3lips package #18 - Set up testing for development efficiency
2. Implement RadarOnlyAssociator for pure radar tracking #20 - Implement RadarOnlyAssociator (CRITICAL)
3. Implement TDOA/FDOA processing for multi-static radar measurements #21 - Add TDOA processing for proper multi-static operation
4. Validate radar-only tracking with synthetic data
5. Consider post-MVP enhancements

The corrected focus is on capability gaps (radar-only operation) rather than optimization of existing working components.