🚗 Travana Ride Booking Process

Complete Technical Flow: Privacy → AI → P2P → Blockchain → Rewards

🔍 Phase 1: Privacy-First Discovery & AI Matching

👤 User Actions
1
📱 Open Travana App
User launches app, wallet auto-connects via Web3 authentication
Tech: MetaMask/WalletConnect integration, JWT token generation
2
📍 Share Fuzzy Location
User allows location access, but only geohash zone is shared (~1.2km area)
Privacy: GPS → Geohash conversion (6-char = tdr5rw)
Actual: 28.6139°N, 77.2090°E
Shared: "tdr5rw" (zone only)
3
🎯 Set Preferences
Destination zone, ride type, max fare, vehicle preferences (EV, luxury, etc.)
AI Input: Preferences stored with zone tags for learning
🧠 System Processing
4
🗄️ Zone Indexing
System indexes user into geohash cluster, no GPS coordinates stored
Database: MongoDB with geospatial indexing on zones
db.users.createIndex({"zone": "2dsphere"})
→ Fast zone-based queries
5
🤖 AI Matching Engine
ML algorithm finds compatible drivers using multi-factor optimization
Algorithm: Zone proximity + price match + preference overlap + historical success
Matching Score = 0.4×zone_dist + 0.3×price_match + 0.2×preferences + 0.1×history
6
💰 Dynamic Pricing
Fair pricing calculated using zone demand, distance estimation, and user willingness
ML Model: XGBoost regression on historical zone-pair pricing

🔗 Phase 2: Secure P2P Connection & Exact Location

👥 Both Users
7
✅ Accept Match
Both rider and driver accept the AI-suggested match
Trigger: Mutual acceptance initiates P2P handshake
8
🔐 P2P Handshake
Direct WebRTC connection established between rider and driver devices
Security: End-to-end encryption, no server intermediary
WebRTC Offer → STUN/TURN → Direct Connection
ENCRYPTED
9
📍 Share Exact Location
Now connected P2P, exact pickup/dropoff coordinates are shared privately
Privacy Win: Travana servers never see exact locations
🛡️ System Security
10
🔄 WebRTC Signaling
Server facilitates initial connection handshake only
Role: Signal server for offer/answer exchange, then disconnects
Socket.io → SDP exchange → ICE candidates → Connection established
11
📊 Minimal Logging
Only ride metadata logged: zone pair, timestamp, fare amount
Privacy: No routes, no exact locations, no conversation data

⛓️ Phase 3: Trustless Payment & Rewards

💳 Payment Flow
12
🔒 Escrow Lock
Rider's payment automatically locked in Aptos smart contract
Blockchain: Aptos Move contract, funds held in escrow
lockFunds(ride_id, amount, rider_wallet, driver_wallet)
13
🚗 Complete Ride
Both parties confirm ride completion via app
Verification: Dual confirmation required for fund release
14
🎁 Receive Rewards
Earn $TVN loyalty points, and ride NFT receipt
Tokenomics: Base ride + bonuses (EV, referral, streak, etc.)
🏦 Smart Contract
15
📜 Contract Execution
Automated escrow contract handles all payment logic
Move Code: Multi-signature release mechanism
public fun releaseFunds(
  ride_id: u64,
  rider_sig: address,
  driver_sig: address
)
16
🎨 NFT Minting
Unique ride receipt NFT created with metadata
Metadata: Zone pair, timestamp, fare, vehicle type, eco-impact
🛡️ Privacy Layer
Geohash Zones: 6-character codes represent ~1.2km² areas
Zero GPS Storage: Exact coordinates never leave user device
P2P Direct: Location sharing happens peer-to-peer only
GDPR COMPLIANT ZERO TRACKING
🧠 AI Matching Layer
Multi-Factor Scoring: Zone + Price + Preferences + History
Dynamic Learning: XGBoost model improves with each ride
Fair Pricing: ML prevents discriminatory surge pricing
BIAS-FREE TRANSPARENT
🔗 P2P Communication Layer
WebRTC Protocol: Direct device-to-device connection
End-to-End Encryption: No server can decrypt messages
Real-time Data: Location, messages, route updates
E2E ENCRYPTED NO MIDDLEMAN
⛓️ Blockchain Settlement Layer
Aptos Network: Fast, low-cost smart contracts
Escrow System: Trustless payment holding & release
NFT Receipts: Verifiable ride history & achievements
TRUSTLESS IMMUTABLE

✅ Complete Flow in Short

Privacy Achieved: Zero location tracking by Travana
Fair Pricing: AI-optimized, transparent fare calculation
Trust Eliminated: Smart contracts handle all payments
Users Rewarded: Tokens, NFTs, and points for participation
Network Effect: Each ride improves matching for everyone