SIH_Fleet_Priority — SIH26123

Edge-AI Based Distributed Fleet Coordination for Autonomous Mobile Robots (AMRs) in Smart Warehouses

A multi-robot warehouse simulation, peer-to-peer coordination protocol, and benchmark harness for decentralized AMR priority and path-conflict resolution.

Quick Start

python3 -m venv .venv
source .venv/bin/activate
python -m pip install -e ".[dev]"
python -m pytest -q
python backend/server.py                       # http://127.0.0.1:8000

Live Deployment: https://bios-sih26123.azurewebsites.net/
Local Dashboard: http://127.0.0.1:8000
Documentation: http://127.0.0.1:8000/docs

Key Features

• Zero third-party dependencies for simulation core and benchmark (stdlib only)
• Decentralized coordination with BIOS_PIBT.6 and Auction V2
• Real edge processes with authenticated UDP multicast
• 3D digital twin dashboard with interactive camera modes
• Comprehensive benchmarking with statistical evidence

Results

✅ 90/90 acceptance runs passed (4, 6, 8 robot fleets)
✅ 0 observed collisions across 88.39 robot-hours
✅ 65.22% / 50.63% / 33.46% minimum task-time reduction vs stop-and-wait

See Benchmark and Evidence for details.

Additional Commands

python edge_demo.py --robots 3 --duration 5    # real processes + signed UDP
python fault_campaign.py --seeds 30 --jobs 8   # loss, partition, crash recovery
python benchmark.py --seeds 30 --jobs 8        # strict SIH acceptance gate
python auction_v2_campaign.py --seeds 30 --jobs 8  # Auction V2 release gate

Documentation

Complete documentation is available at /docs when running the server:

• Problem Statement & Requirements
• Architecture & Protocol Design
• Benchmark Results & Evidence
• Deployment Guide
• API Reference
• Testing & Validation

Team: Bharat Electronics Limited · Software · Robotics and Drones
Problem Statement: PS26123