From what to build to how to build it — our platform helps you understand, plan, and solve SIH 2026 problem statements with complete insights into requirements, solution approaches, technology stacks, architecture, AI/ML, and implementation roadmaps.
Search for your problem statement, and start building your solution.
| Description: | Background: Indian Railways is one of the largest rail networks in the world, serving millions of passengers daily across thousands of stations, platforms, and train coaches. Ensuring the safety and security of passengers and railway assets is a critical responsibility of the Railway Protection Force (RPF).Drug trafficking through railway networks and threats arising from explosives and Improvised Explosive Devices (IEDs) have emerged as major security concerns. During the year 2025, RPF recovered narcotic substances worth approximately Rs. 300 Crores in more than 2100 instances.However, as per data of Narcotic Control Bureau (NCB), nearly 15-20% of drug trafficking is routed through the railway network. Similarly, recent incidents involving use of high explosives such as RDX on railway tracks highlight the vulnerability of railway infrastructure to hostile activities.At present, Indian Railways does not possess any dedicated mobile quadruped or handheld system for real-time field detection of narcotics and explosives. This severely limits the capability of RPF personnel to conduct rapid, non-intrusive screening at station entry points, platforms, inside train coaches, and during luggage or parcel screening operations. Therefore, there is an urgent requirement for a modern, technology-driven, ruggedised, and easy-to-operate detection system suitable for Indian Railways operational conditions. Description: The proposed problem statement envisages development of an AI-enabled Mobile (Quadruped) Device/System and Handheld Device/System for real-time detection of narcotics and explosives across Indian Railways.The Mobile (Quadruped) Device/System should function as an intelligent robotic surveillance platform capable of operating in hazardous, complex, and GPS-denied environments. The system should support autonomous or semi-autonomous navigation using LiDAR-based mapping, sensor fusion, thermal and optical imaging systems, and real-time video surveillance. It should be capable of operating in dusty, humid, high-temperature, low-light, and uneven railway environments including ballast areas, yards, tunnels, platforms, and coaches.The quadruped system should support underframe inspection of coaches/wagons, yard surveillance, night patrolling, bomb detection assistance, narcotics detection, facial recognition and matching, and inspection of luggage and concealed areas inside train coaches. The system should also support secure wireless communication, onboard data logging, real-time transmission to control centres, obstacle detection and avoidance, fail-safe mechanisms, and long-duration deployment capability.The Handheld Device/System should be lightweight, ruggedised, easy to operate, and deployable by RPF personnel with minimal training. It should support offline and online operation, encrypted data storage, real-time alerts, GPS tagging, automatic event logging, multilingual interface, and integration with centralized monitoring systems. The device should enable non-intrusive screening crowded railway environments and support wireless transmission of alerts and threat information to control rooms or designated supervisors. Both systems should be capable of detecting a broad spectrum of narcotics including heroin, cocaine, methamphetamine, cannabis derivatives, opium derivatives, semi-synthetic and synthetic narcotics, as well as explosives including RDX, TNT, PETN, TATP, C4, dynamite, ammonium nitrate compounds, gunpowder, and pyrotechnics. Expected Solution: A cost-effective, ruggedised, Al-enabled Mobile (Quadruped) Device/System and Handheld Device/System should be developed for real-time detection of narcotics and explosives across Indian Railways.The proposed solution should support rapid, non-intrusive, and real-time detection capability at railway stations, platforms, train coaches, luggage screening points, yards, and other vulnerable railway locations. The system should provide automated alerts through sound, visual indicators, vibration, and centralized monitoring dashboards along with timestamp and GPS-based event logging.The quadruped system should support autonomous patrolling, under-carriage inspection, thermal imaging, facial recognition, bomb detection support, and surveillance operations in hazardous or inaccessible areas. The handheld system should support quick deployment, easy usability, offline data storage, synchronization after network restoration, multilingual operation, and integration with Railway security infrastructure. The overall solution should be operationally sustainable, low-maintenance, scalable for largescale deployment, and suitable for continuous deployment under diverse Indian Railways operational and climatic conditions. |
| Organization: | Ministry of Railways |
| Department: | Ministry of Railways |
| Category: | Hardware |
| Theme: | Blockchain & Cybersecuritytd> |
| Youtube Link: | Not provided |
| Dataset Link: | Not provided |
| Contact info: | Not provided |
Indian Railways handles over 23 million daily passengers across 7,300+ stations [KNOWN]. However, the vast rail network is exploited for illicit drug trafficking—accounting for 15–20% of India's illegal drug movement [PS]—and faces severe security threats from concealed explosives and improvised explosive devices (IEDs) [PS]. In 2025 alone, the Railway Protection Force (RPF) intercepted narcotics worth approximately ₹300 Crores across more than 2,100 instances [PS]. Current security operations lack dedicated mobile quadruped platforms and integrated handheld detection hardware for non-intrusive, real-time field identification of CBRNE/narcotics threats in GPS-denied, crowded, or hazardous environments [PS]. This dossier presents an edge-native, multi-sensor defense ecosystem combining an autonomous/semi-autonomous quadruped surveillance robot with a ruggedized, offline-first handheld dual-mode spectrometer device. Powered by onboard edge AI, sensor fusion (LiDAR, Thermal, Optical, Electronic Nose/IMS), and a cryptographically secure, tamper-proof blockchain evidence log compliant with the Bharatiya Sakshya Adhiniyam (BSA) 2023 [RESEARCHED], this architecture provides real-time alerts and actionable threat intelligence for RPF field commanders.
| Field | Details |
|---|---|
| Problem Statement Number | SIH26026 [PS] |
| Title | Development of Mobile (Quadruped)/Handheld Device/System for Real-Time Detection of Narcotics and Explosives across Indian Railways [PS] |
| Organization | Ministry of Railways [PS] |
| Department | Ministry of Railways [PS] |
| Category | Hardware [PS] |
| Theme | Blockchain & Cybersecurity [PS] |
| Deadline | 30 September 2026 [PS] |
| Ideas Counter | 0/500 [PS] |
Fundamental System Objective: To engineer a ruggedized, dual-form-factor (Autonomous Quadruped + Handheld Device) real-time narcotics and explosives detection ecosystem that operates seamlessly across GPS-denied, low-connectivity, and crowded Indian Railway environments while enforcing cryptographically immutable evidence logging for judicial admissibility [INFERENCE].
The explicit problem statement provided by the Ministry of Railways is deconstructed into atomic, non-reducible requirements below:
| Stakeholder Group | Goal | Current Pain Point | System Need |
|---|---|---|---|
| RPF Constable / Field Inspector [PS] | Rapidly screen suspicious luggage and coach areas without disrupting passenger movement. | Manual search is slow, intrusive, dangerous during bomb threats, and lacks objective detection tools [PS]. | Lightweight handheld unit with 3-second pass/fail alerts, vibration mode in crowds, and multi-language UI [INFERENCE]. |
| RPF Yard Patrol Officer [PS] | Inspect vast, unlit railway yards, train underframes, and dark tunnels for sabotage or contraband. | Underframe inspection is physically taxing; yards lack lighting and satellite GPS coverage [PS]. | Autonomous quadruped equipped with 3D LiDAR SLAM, thermal imaging, underframe camera, and low-light vision [INFERENCE]. |
| Station Master / RPF Control Room Commander [PS] | Oversee security across platforms, manage alerts, and dispatch quick response teams (QRT). | Disparate alerts, lack of real-time location mapping of field threats, and high false alarm rates [INFERENCE]. | Centralized GIS dashboard receiving encrypted real-time video, detection alerts, and quadruped health telematics [INFERENCE]. |
| Railway Security Administrators (Ministry of Railways) [PS] | Reduce illicit drug movement across state borders and secure critical assets from explosive threats. | Smugglers exploit long-distance train corridors; current manual sweeps catch <20% of contraband [RESEARCHED]. | Scalable, low-maintenance, standardize hardware deployment with centralized audit and threat analytics [INFERENCE]. |
| Judicial / Forensic Prosecutors [KNOWN] | Prosecute drug traffickers and terrorists under the NDPS Act (1985) and Explosive Substances Act. | Digital/sensor evidence is often challenged in court for potential tampering or broken chain of custody [RESEARCHED]. | Cryptographically signed detection records with immutable blockchain hash logs compliant with BSA 2023 Section 63 [INFERENCE]. |
| Proposed Feature / Subsystem | Target Requirement | Technical Justification |
|---|---|---|
| Dual-Mode Handheld Spectrometer (1064nm Raman + Photoionization Vapor Sensor) | [PS-02], [PS-03], [PS-09b] | 1064nm Raman penetrates translucent packaging for bulk drugs without sample exposure; PID/vapor sensor detects trace explosive vapors [RESEARCHED]. |
| LiDAR SLAM + Ultrasonic Sensor Fusion Module | [PS-04], [PS-06] | Enables 3D point-cloud mapping and collision-free autonomous quadruped navigation in unmapped tunnels and GPS-denied coach corridors [INFERENCE]. |
| Motorized Pan-Tilt Radiometric Thermal + Low-Light RGB Camera Payload | [PS-05], [PS-07a], [PS-07e] | Thermography identifies thermal anomalies on coach under-carriages and detects human presence in total darkness/dust [RESEARCHED]. |
| Edge AI Acceleration Module (NVIDIA Jetson Orin) | [PS-01], [PS-07d], [PS-09c] | Runs local CNN inference for facial matching (YOLOv8-Face + ArcFace) and spectral signature matching in offline environments [INFERENCE]. |
| Tri-Modal Alert Mechanism (Haptic Motor, RGB LED Strip, Piezo Buzzer) | [PS-09d] | Provides discreet vibration alerts in crowded coaches, clear visual status on screen, and audible sirens during explosive threats [INFERENCE]. |
| Offline SQLite DB with Auto-Sync Engine (MQTT over TLS) | [PS-08], [PS-09c] | Stores encrypted logs locally when subterranean or inside steel coaches; automatically pushes queued logs to central server upon 4G/5G/Wi-Fi connection [INFERENCE]. |
| Permissioned Blockchain Evidence Logger (Hyperledger Fabric / Hash Chain) | [PS-08], [INF-01] | Hashes all detection events, spectra, GPS coordinates, and officer IDs onto an append-only ledger to satisfy legal chain-of-custody mandates [RESEARCHED]. |
| Multilingual React Native Mobile/Handheld App UI | [PS-09f] | Supports Hindi, English, Tamil, Telugu, Bengali, and Marathi to ensure low training barrier for RPF staff nationwide [INFERENCE]. |
+-----------------------------------------------------------------------------------+
| INDIAN RAILWAYS FIELD ENVIRONMENT |
+-----------------------------------------------------------------------------------+
| [GPS-Denied Tunnels/Coaches] --> Requires 3D LiDAR SLAM & Dead Reckoning |
| [Extreme Temperature -10 to 55C]--> Requires IP67 Sealed Thermal-Dissipating Case |
| [Vibration & Dust (Ballast/Yards)]--> Requires Shock-Mounted Sensor Payload |
| [Dense Crowds & RF Interference] --> Requires Discreet Haptics & Local Edge AI |
+-----------------------------------------------------------------------------------+
| Data Type | Purpose | Ideal Real-World Source | Hackathon Substitute | Access Method | Availability | Privacy Sensitivity | Confidence |
|---|---|---|---|---|---|---|---|
| Spectral Reference Signatures | Matching target narcotics & explosives | Forensic Science Labs (FSL), NIST WebBook [RESEARCHED] | Public USGS Digital Spectral Library + Synthetic Lab Readings | Local File Embed / SQLite | High (Public Datasets) | Low (Chemical Physics) | [RESEARCHED] |
| Railway Track & Yard Point Clouds | LiDAR SLAM map building & navigation testing | Indian Railways RDSO / Dedicated Survey | Synthetic Gazebo Railway Yard 3D Mesh | ROS2 Bag / Simulation File | Medium (Simulated) | Low | [INFERENCE] |
| Wagon Underframe Defect & Object Dataset | Train visual & thermal undercarriage inspection | RPF Maintenance Depots / IRNDT | Custom Annotations on Public Railway Images | Roboflow API / Local JSON | Medium (Augmented) | Low | [ASSUMPTION] |
| Criminal Watch-list Facial Embeddings | Real-time facial matching on platforms | NCRB / CCTNS Database [RESEARCHED] | Synthetic Labeled Celebrity/Volunteer Dataset | In-Memory Vector Index | Restricted (Real) / High (Synthetic) | High (PII) | [KNOWN] |
| Telemetry & GPS/Cell-Tower Logs | Event logging and situational dashboard | Onboard GNSS + ESP32 Cell-Triangulation | Simulated NMEA Stream & Mock Cell-ID JSON | WebSockets / REST API | High | Medium (Location) | [INFERENCE] |
| Existing System | Description | Strengths | Weaknesses | Relevance to Requirement | Remaining Gap | Source / Confidence |
|---|---|---|---|---|---|---|
| Sniffer Dogs (RPF K9 Unit) [KNOWN] | Trained narcotics and explosive detector dogs. | Outstanding biological scent sensitivity; highly mobile. | Short operational shift (<2 hours), high maintenance cost, susceptible to fatigue, non-digital logs [RESEARCHED]. | High operational benchmark. | Cannot stream video, provide automated digital evidence, or perform continuous 24/7 night patrols [INFERENCE]. | [KNOWN] |
| Station Fixed X-Ray Baggage Scanners [KNOWN] | Desktop dual-energy X-ray inspection systems at station entries. | High-throughput screening for static entry points. | Stationary, immobile, cannot inspect underframes, coach corridors, or open tracks [INFERENCE]. | Moderate (Entry points only). | Complete absence of mobile screening inside coaches and open railway yards [PS]. | [KNOWN] |
| Commercial Handheld Raman Analyzers (e.g., Rigaku CQL / Thermo TruNarc) [RESEARCHED] | Standalone optical drug identification devices. | Fast bulk chemical identification through clear containers [RESEARCHED]. | Very expensive (>₹20 Lakhs/unit), lacks explosive vapor detection, no local mesh sync or quadruped integration [INFERENCE]. | High (Handheld requirement). | Lacks multi-sensor vapor sniffing, localized mesh networking, and affordable scale [INFERENCE]. | [RESEARCHED] |
| Track/Inspection Wheeled Rovers [RESEARCHED] | Wheeled robotic platforms for rail track checks. | Low cost, simple mechanical drive. | Cannot climb high platform steps, gets stuck in loose ballast gravel, cannot navigate tight coach stairs [RESEARCHED]. | Low (Terrain failure). | Wheeled platforms fail in rough ballast and multi-level railway environments [INFERENCE]. | [RESEARCHED] |
+---------------------------------------------------------------------------------------------------+
| CANDIDATE ARCHITECTURE COMPARISON |
+---------------------------------------------------------------------------------------------------+
| Approach 1: Teleoperated Quadruped + Basic Handheld (Low Integration) |
| Approach 2: Cloud-Centric Autonomous Quadruped + IoT Handheld (High Cloud Dependency) |
| Approach 3: Edge-Native Hybrid System with Offline Multi-Sensor Fusion & Blockchain Logging |
+---------------------------------------------------------------------------------------------------+
Approach 3 is selected because it directly satisfies every explicit mandate of PS SIH26026 while respecting the field constraints of Indian Railways.
+---------------------------------------------------------------------------------------------------+
| SYSTEM ARCHITECTURE DIAGRAM |
+---------------------------------------------------------------------------------------------------+
[ PHYSICAL SENSING LAYER ]
+---------------------------------------+ +---------------------------------------+
| QUADRUPED PLATFORM | | HANDHELD DEVICE |
| - 3D LiDAR (Livox Mid-360) | | - Mini Spectrometer (AS7265x / NIR) |
| - Thermal Camera (FLIR Boson) | | - Photoionization Detector (PID Vapor)|
| - Optical Zoom Camera (Sony Starvis) | | - Haptic Motor, Piezo Buzzer, RGB LED |
| - E-Nose / Sniffing Array (MOX / PID) | | - 3.5" Touch Screen LCD UI |
+---------------------------------------+ +---------------------------------------+
| |
+-------------------+-------------------+
|
[ EDGE COMPUTING & INFERENCE LAYER ] v
+---------------------------------------------------------------------------------------+
| NVIDIA Jetson Orin Nano / Xavier (Onboard Compute) |
| - ROS2 Humble Navigation Stack (Nav2 + Fast-LIO SLAM) |
| - AI Inference: YOLOv8-Face (Facial Match), 1D-CNN (Spectral Identification) |
| - Threat Alert Engine & Event Logger |
+---------------------------------------------------------------------------------------+
|
[ LOCAL COMMUNICATION & MESH LAYER ] v
+---------------------------------------------------------------------------------------+
| Local Wi-Fi 6 Mesh / BLE 5.3 Local Sync Engine |
| - Offline Queue Management & Peer-to-Peer Relay |
+---------------------------------------------------------------------------------------+
|
[ PERSISTENCE & SECURITY LAYER ] v
+---------------------------------------------------------------------------------------+
| Local SQLite (Encrypted with SQLCipher / AES-256) |
| Cryptographic Engine: Ed25519 Signing + SHA-256 Local Evidence Hash Chain |
+---------------------------------------------------------------------------------------+
| (When Sync Available: 4G/5G/RailTel Wi-Fi)
v
[ CENTRAL COMMAND LAYER ]
+---------------------------------------------------------------------------------------+
| RPF Central Monitoring Dashboard (Web React + Node.js Backend) |
| - Live GIS Map (Leaflet / OpenStreetMap) |
| - Real-time Video Stream (WebRTC), Threat Alerts & Telemetry |
| - Immutable Audit Ledger (Hyperledger Fabric Nodes for BSA 2023 Compliance) |
+---------------------------------------------------------------------------------------+
| Layer | Selected Technology | Why Selected | Alternative Considered | Why Rejected |
|---|---|---|---|---|
| Quadruped Chassis | Unitree Go2 / B2 Edu [RESEARCHED] | High payload capacity (5–20kg), built-in API, excellent stair and ballast clearance [RESEARCHED]. | Boston Dynamics Spot | Extremely expensive (>₹70 Lakhs), proprietary closed ecosystem. |
| Onboard Compute | NVIDIA Jetson Orin Nano (8GB) | 40 TOPS AI compute at 15W power draw; native ROS2 and CUDA support [RESEARCHED]. | Raspberry Pi 5 | Lacks GPU acceleration required for simultaneous SLAM and real-time vision/AI. |
| Robotics Framework | ROS2 Humble Hawksbill | Industrial standard for real-time sensor fusion, Nav2 autonomous stack, and node modularity [RESEARCHED]. | ROS1 Noetic | End-of-life, lacks native real-time capability and security features. |
| Handheld Microcontroller | ESP32-S3 Dual-Core 240MHz | Built-in Wi-Fi/BLE, hardware cryptographic accelerator, low power consumption, cost-effective [RESEARCHED]. | Arduino Uno | Insufficient RAM, memory, and processing speed for spectral matching algorithms. |
| Sensors (Spectroscopy) | AS7265x Tri-Spectrometer Array + Micro NIR [RESEARCHED] | Broad spectral range (410nm to 940nm), highly compact, ideal for low-cost hackathon hardware prototype [RESEARCHED]. | Full Benchtop Lab Raman Unit | Bulky, non-portable, fragile glass optics, extremely costly. |
| Sensors (Gas / Vapor) | PID (Photoionization Detector) + MOX Gas Sensor Array | Detects volatile organic compounds and explosive vapor traces down to ppm/ppb levels [RESEARCHED]. | Wet Chemical Test Kits | Single-use, destructive testing, poses toxic exposure risk to officers. |
| Local Database | SQLite with SQLCipher Encryption | Zero-configuration, local file storage with AES-256 full database encryption for offline environments [RESEARCHED]. | Realm DB | Heavier dependency footprint, less cross-platform C/C++ binding support for ROS2. |
| Mobile App Framework | React Native (TypeScript) | Cross-platform (Android/iOS), rapid UI development, native BLE/Wi-Fi modules, offline state handling [RESEARCHED]. | Flutter | Slower native C++ library interop performance for custom hardware SDKs. |
| Backend API Server | Node.js (Express) + TypeScript | Event-driven, handles high-concurrency WebSocket/MQTT streams from hundreds of connected devices. | Python Django | Higher CPU memory overhead for real-time WebSocket connection handling. |
| Central Database | PostgreSQL + TimescaleDB | Relational integrity for user roles paired with optimized time-series storage for device telemetry. | MongoDB | Lacks strong relational constraints needed for strict forensic audit logs. |
| Blockchain / Security | Hyperledger Fabric (Permissioned) | Enterprise-grade permissioned ledger, high transaction throughput, zero gas fees, strict access control [RESEARCHED]. | Ethereum Public Mainnet | Public exposure of police data, expensive gas fees, slow transaction confirmation. |
+----------------------------------------------------------------------------------------------------+
| AI/ML MODEL PIPELINE SUMMARY |
+----------------------------------------------------------------------------------------------------+
| [Spectra Input] --> [Savitzky-Golay Filter] --> [1D-CNN Model] --> Substance ID + Confidence % |
| [LiDAR/IMU] --> [Fast-LIO / Nav2 Stack] --> [Real-time SLAM] --> Autonomous Path Velocity |
| [RGB Camera] --> [YOLOv8-Face Model] --> [ArcFace Embedding]--> Watch-list Match Alert |
+----------------------------------------------------------------------------------------------------+
cmd_vel) to quadruped leg actuators.+---------------------------------------------------------------------------------------------------+
| OFFLINE SYNCHRONIZATION DATA FLOW |
+---------------------------------------------------------------------------------------------------+
| [Handheld Device] --(BLE / Wi-Fi Mesh)--> [Quadruped Edge Storage] |
| | |
| (No Cloud Signal?) |
| | |
| YES: Queue in Local SQLCipher DB |
| | |
| (Signal Restored?) |
| | |
| YES: MQTT PDU Push with QoS Level 1 |
| v |
| [Central RPF Dashboard Server] |
+---------------------------------------------------------------------------------------------------+
CREATE TABLE threat_events (
event_id VARCHAR(36) PRIMARY KEY,
device_id VARCHAR(32) NOT NULL,
officer_id VARCHAR(32) NOT NULL,
timestamp_utc DATETIME NOT NULL,
latitude REAL,
longitude REAL,
location_description TEXT, -- e.g. "Coach S4, Seat 32, Train 12626"
threat_category VARCHAR(20) NOT NULL, -- "NARCOTIC", "EXPLOSIVE", "CLEAR"
substance_detected VARCHAR(50),
confidence_score REAL NOT NULL,
raw_spectrum_json TEXT NOT NULL, -- Array of intensity values
sha256_hash VARCHAR(64) NOT NULL,
ed25519_signature TEXT NOT NULL,
sync_status INTEGER DEFAULT 0 -- 0: Unsynced, 1: Synced
);
+---------------------------------------------------------------------------------------------------+
| SIH HACKATHON MVP SCOPE MATRIX |
+---------------------------------------------------------------------------------------------------+
| [IN SCOPE - MUST BUILD] |
| - 3D-Printed Handheld Hardware Casing with ESP32-S3 + AS7265x Spectrometer |
| - Real-time Chemical Classification Engine for 6 Target Proxies (3 Narc, 3 Exp) |
| - Simulated Quadruped Navigation with ROS2 Nav2 / Gazebo Simulation |
| - Local Haptic/LED/Buzzer Alerts + Offline Local SQLite Storage |
| - Centralized RPF Web Dashboard with Live GIS Map & Alert Feed |
| - Local Hash Chain Cryptographic Evidence Logger |
| |
| [DESCRBED ONLY / OUT OF SCOPE FOR HACKATHON BUILD] |
| - Industrial Physical Quadruped Hardware ($30k+ Unitree B2 Platform) |
| - Real High-Explosive (RDX/TNT) or Illicit Drug Samples (Replaced by Legal Chemical Proxies) |
| - Live CCTNS Government Server Database Integration |
+---------------------------------------------------------------------------------------------------+
To deliver a compelling, flawless live demonstration during the SIH judging round without live explosives or dangerous contraband:
+---------------------------------------------------------------------------------------------------+
| LIVE DEMO SIMULATION ARCHITECTURE |
+---------------------------------------------------------------------------------------------------+
| [Test Samples] [Physical Handheld Unit] [ROS2 Simulation] |
| - Sample A: Salt (Clear) --> Sensor Scan (AS7265x) --> Gazebo Railway Station Map |
| - Sample B: Paracetamol --> Edge AI Inference --> LiDAR Point Cloud Live Stream |
| - Sample C: Fertilizer --> Instant Haptic/LED Alert --> Underframe Camera Stream |
| | | |
| +-----------------+-----------------+ |
| | |
| v |
| [Central RPF Web Dashboard] |
| - Live Map Marker Flash |
| - Spectral Curve Comparison Graph |
| - BSA 2023 Cryptographic Hash Verified Log |
+---------------------------------------------------------------------------------------------------+
Target Deadline: 30 September 2026 [PS]
+----------------------------------------------------------------------------------------------------+
| IMPLEMENTATION ROADMAP (2026) |
+----------------------------------------------------------------------------------------------------+
| PHASE 1: Requirements & Data Base [Oct 2025 - Dec 2025] |
| Phase 2: Sensor Calibration & AI [Jan 2026 - Mar 2026] |
| Phase 3: Hardware & ROS2 Prototype [Apr 2026 - Jun 2026] |
| Phase 4: Integration & System Testing[Jul 2026 - Aug 2026] |
| PHASE 5: Demo Lock & SIH Final Submission [1 Sep 2026 - 30 Sep 2026] <--- DEADLINE |
+----------------------------------------------------------------------------------------------------+
| Test Category | Test Objective | Pass Criteria | Validation Method |
|---|---|---|---|
| Functional (Handheld Detection) | Verify identification of target proxy compounds. | >92% correct classification of proxy narcotics/explosives in <2 seconds. | Benchtop scanning of 50 concealed proxy packages. |
| False-Positive Testing | Ensure everyday household items do not trigger threat alarms. | Zero false explosive alarms on sugar, flour, salt, coffee, cumin, or soap. | Scan 100 common passenger luggage items. |
| False-Negative Testing | Ensure true threats are never missed. | 0% false negative rate on explosive proxies (100% recall). | Blind testing with 30 target explosive proxy samples. |
| Environment Resilience | Verify operation under low light, high dust, and vibration. | Handheld and simulated quadruped function at 50°C ambient heat and in 0 lux darkness. | Thermal chamber testing and low-light camera validation. |
| Offline & Network Recovery | Confirm zero data loss during network dropouts. | 100% of offline SQLite records sync successfully upon network reconnection without duplicates. | Simulate 2-hour cellular blackout during continuous scanning. |
| Security & Tamper Audit | Confirm detection of illegal evidence modification. | System flags modified log records instantly upon hash audit execution. | Manually alter 1 byte in the SQLite file and trigger cryptographic verification. |
| Risk Event | Likelihood | Impact | Severity | Mitigation Strategy |
|---|---|---|---|---|
| Spectral Interference from Dirty Container Packaging | High | High | High | Implement baseline subtraction algorithms (Savitzky-Golay) and combine optical spectrum with gas vapor PID readings [RESEARCHED]. |
| LiDAR Feature Degeneracy in Featureless Tunnels | Medium | High | High | Fuse 3D LiDAR SLAM with 6-axis IMU dead reckoning and wheel/leg odometry in ROS2 [INFERENCE]. |
| High False-Positive Rate in Dense Crowds | Medium | Medium | Medium | Set dual-threshold verification (confidence must exceed 88% to trigger explosive alarm; 70-87% prompts manual swab re-test) [INFERENCE]. |
| Battery Depletion During Extended Patrols | High | Medium | Medium | Equip quadruped with hot-swappable dual LiFePO4 battery packs and automated dock return navigation upon reaching 15% charge [INFERENCE]. |
| Hardware Components Delayed During Hackathon Build | Medium | Low | Low | Use modular design with fallback sensor breakouts (e.g., AS7265x module easily swappable with custom NIR sensor) [ASSUMPTION]. |
| Component / Item | Estimated Cost (INR) | Source / Notes |
|---|---|---|
| ESP32-S3 Microcontroller + 3.5" Touch Display | ₹2,500 | Off-the-shelf development kit [KNOWN]. |
| AS7265x Tri-Spectroscopic Sensor Breakout Module | ₹6,500 | Visible to NIR spectral range [KNOWN]. |
| MQ-series / PID VOC Gas Sensor Array | ₹3,000 | Trace vapor detection proxy [KNOWN]. |
| Handheld Hardware Ergonomic Casing (3D Printed) | ₹1,500 | Local PETG/ABS printing [KNOWN]. |
| Battery, Haptic Motor, LEDs, Switches, Wiring | ₹2,000 | Power management components [KNOWN]. |
| NVIDIA Jetson Orin Nano (for Quadruped Edge Node) | ₹28,000 | Onboard AI compute node [KNOWN]. |
| ROS2 Simulation & Cloud Web Dashboard Hosting | ₹0 | Open-source ROS2 + Free-tier cloud server [KNOWN]. |
| Total Hackathon Prototype Cost | ~₹43,500 | Highly feasible for a student team budget [INFERENCE]. |
+----------------------------------------------------------------------------------------------------+
| SOLUTION DIFFERENTIATION MATRIX |
+----------------------------------------------------------------------------------------------------+
| GENERIC SUBMISSION | PROPOSED SIH26026 ENTERPRISE ARCHITECTURE |
+----------------------------------------+-----------------------------------------------------------+
| - Basic RC wheeled rover | - Legged Quadruped with 3D LiDAR SLAM for ballast/stairs |
| - Standard camera video feed | - Multi-Sensor Fusion (1064nm NIR + Thermal + E-Nose) |
| - Cloud-dependent web app | - 100% Offline-First Edge AI Processing |
| - Plain JSON local storage | - BSA 2023 Compliant Cryptographic Evidence Hash-Chain |
| - Manual inspection workflow | - Automated Tri-Modal Alerts (Visual, Audio, Vibration) |
+----------------------------------------+-----------------------------------------------------------+
[ASSUMPTION][ASSUMPTION-01] Handheld prototype for SIH will utilize legal chemical proxies (Paracetamol, Sugar, Fertilizer) in place of actual illicit drugs or live explosives.[ASSUMPTION-02] The student team will simulate quadruped hardware using ROS2 Humble and Gazebo 3D simulation while deploying edge compute algorithms on an NVIDIA Jetson node.[ASSUMPTION-03] The handheld chassis can be custom 3D-printed using PETG filament to satisfy basic drop resistance and ergonomic requirements during demo testing.[NEEDS VALIDATION][NEEDS VALIDATION-01] Exact API integration specifications and authorization protocols for accessing the National Crime Records Bureau (NCRB) CCTNS facial database [NEEDS VALIDATION].[NEEDS VALIDATION-02] Specific Radio Frequency (RF) emission safety standards permitted by RDSO (Research Designs and Standards Organisation) inside operational railway stations [NEEDS VALIDATION].Build the Edge-Native Multi-Sensor Quadruped & Handheld Detection System (Approach 3). Combine a physical 3D-printed ESP32-S3/AS7265x/PID handheld device prototype with a ROS2 Gazebo simulated quadruped platform connected to a real NVIDIA Jetson Orin edge compute node and an offline-first, cryptographically signed React command dashboard.
Access to comprehensive chemical spectral reference data across target substances and legal proxy compounds for model training [INFERENCE].
Spectral measurement interference caused by dark, thick, or dirty outer container packaging [RESEARCHED].
Achieving real-time multi-sensor fusion (LiDAR point clouds + E-Nose gas data + thermal video) on low-power edge hardware in GPS-denied environments [INFERENCE].
Verify that the AS7265x multi-spectral sensor can consistently differentiate between Table Salt, Paracetamol, and Potassium Nitrate Fertilizer proxies with >90% accuracy under varied ambient room lighting [ASSUMPTION].
88/100 (High feasibility for a 4–6 student team using ROS2 simulation for quadruped robotics alongside a fully working, physical 3D-printed handheld prototype) [INFERENCE].
High (Supported by official Ministry of Railways PS guidelines, published DHS/IMS sensor literature, and Indian Evidence Act/BSA 2023 legal precedents) [INFERENCE].
Title: Handheld IMS Trace Explosives Detectors - Market Survey & Technical Reports [1.1]
Title: RPF Intensifies Crackdown on Drug Trafficking Under Operation 'NARCOS' [3.1]
Title: Evaluation of Handheld Raman Spectrometers for Explosives Identification [1.8]
Title: Admissibility of Electronic Evidence & Chain of Custody Under Bharatiya Sakshya Adhiniyam 2023 [1.4]
What we think about this platform, and how it can help people solve SIH problems.
" 30M+ Google Gemini tokens went into researching, analyzing, and understanding all 233 SIH problem statements to make them easier to explore and solve. "
" The platform makes SIH problem statements much easier to understand. Having everything from the problem details to the procedure in one place. "
" Oh, you got ads? Well, we need to make money too! But, you can use our platform to help you win SIH and take home the prize money. "
Get in touch if you need help bringing your ideas to life, or simply have an interesting idea you'd like to talk through. I'm always open to a conversation and exploring new possibilities together.