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: Jharkhand is lndia's leading mineral-producing state, with coal mines, steel plants, and mica processing units employing hundreds of thousands of workers many of them young tribal recruits with no prior industrial exposure. Classroom-based safety training using static manuals has documented retention rates below 20% after one week. Live drills are operationally disruptive, and VR headset simulators are inaccessible to small-scale mines and contract workers.. The DGMS, Dhanbad, recorded 48 fatal mine accidents in Jharkhand in 2022-23, a large share involving workers with under 30 days of orientation. The Factories Act, 1948 and Mines Act, 1952 mandate periodic safety certification, yet no standardised digital training platform exists in regional languages, and physical certificates have no mechanism to verify comprehension. Description: Design and develop a mobile AR-based vocational training and safety certification platform running on mid-range Android smartphones (Android 10+, no external headset required), accessible to workers across Jharkhand's mining, steel, and mica sectors. The platform must deliver interactive AR training modules covering five industrial safety domains: (l) Fire & Explosion Response-exit identification, extinguisher use, and evacuation sequencing overlaid on real surroundings via phone camera; (2) Gas Leak & Confined Space Protocol-hazard zone recognition, PPE selection, and buddy-system procedures simulated in AR; (3) Machinery. Expected Solution: A working Android APK demonstrating at least two complete AR training modules, an assessment engine, QR-based certificate generation and verification, Hindi and Santali localisation, offline functionality, and a web admin compliance dashboard- submitted with a demo video and public GitHub repository. |
| Organization: | Government of Jharkhand |
| Department: | Department of Higher & Technical Education |
| Category: | Software |
| Theme: | Smart Educationtd> |
| Youtube Link: | Not provided |
| Dataset Link: | Not provided |
| Contact info: | Not provided |
The Government of Jharkhand faces a critical occupational safety challenge in its mining and manufacturing belts (Dhanbad, East/West Singhbhum, Ranchi). Hundreds of thousands of workers—many newly recruited tribal contract workers—enter high-risk underground coal mines, steel plants, and mica processing units with minimal practical safety orientation. Static classroom manuals yield under 20% knowledge retention after one week, while VR headsets remain cost-prohibitive and impractical for remote small-scale operations. Directorate General of Mines Safety (DGMS) records show 48 fatal accidents in Jharkhand in 2022–23, predominantly involving workers in their first 30 days. To eliminate this gap, we propose SurakshaAR, a lightweight, offline-first mobile Augmented Reality (AR) safety training and verification platform. Running on mid-range Android 10+ smartphones without external headsets, SurakshaAR overlays interactive safety protocols—such as fire extinguisher operation and gas leak response—onto the worker’s physical environment in Hindi and Santali (Ol Chiki script). It pairs gamified AR simulation with a tamper-proof, offline-verifiable QR certificate system and a central web compliance dashboard for mine management and safety inspectors under the Mines Act, 1952 and Factories Act, 1948.
| Parameter | Value |
|---|---|
| Problem Statement ID | SIH26041 [PS] |
| Title | AR-Based Vocational Training Simulator for Industrial Safety in Jharkhand's Mining & Manufacturing Sector [PS] |
| Organization | Government of Jharkhand [PS] |
| Department | Department of Higher & Technical Education [PS] |
| Category | Software [PS] |
| Theme | Smart Education [PS] |
| Submission Deadline | 30 September 2026 [PS] |
| Ideas Counter | 0 / 500 [PS] |
| Target OS / Hardware | Android 10+ (API Level 29+), Mid-Range Smartphones, No external headset [PS] |
| Key Deliverables | Android APK (2+ complete AR modules), Assessment Engine, QR Certificate System, Hindi/Santali Localisation, Offline Mode, Web Compliance Dashboard, Demo Video, Public GitHub Repository [PS] |
To deliver an accessible, offline-capable mobile AR safety training and tamper-proof certification platform that converts static safety manuals into interactive, multi-lingual spatial simulations on standard smartphones, dramatically reducing early-tenure industrial fatalities across Jharkhand.
The explicit problem statement supplied in the SIH JSON is broken down into atomic requirements below. Each requirement is mapped to its source text, domain interpretation, implementation implications, and confidence rating.
Noto Sans Ol Chiki) [RESEARCHED] and pre-rendered audio voiceover files in the APK assets to ensure Ol Chiki rendering and spoken guidance work seamlessly offline without system font dependency.| Stakeholder Group | Goal | Current Pain Point | System Need |
|---|---|---|---|
| Contract & Tribal Mine Workers | Perform hazardous mining/factory jobs safely without injury; achieve mandatory safety certification. [KNOWN] | Static classroom manuals yield $<20%$ retention [PS]; language barrier with English/Hindi materials; high fatigue; fear of complex machinery. | Interactive, visual AR learning on personal/shared smartphones in native Santali (Ol Chiki) & Hindi with zero network reliance. [INFERENCE] |
| Shift Safety Officers / Mine Managers | Ensure $100%$ workforce safety orientation compliance under Mines Act 1952 / CMR 2017; lower site accidents. [RESEARCHED] [1.4] | Physical paper certificates are easily forged and don't prove real hazard handling comprehension [PS]; live drills halt mine production. [PS] | Offline QR scanner app to instantly verify worker safety certification and real practical assessment scores at mine gate. [INFERENCE] |
| DGMS & State DISH Inspectors | Audit mine/factory safety compliance; investigate accident causes; enforce statutory training mandates. [RESEARCHED] [1.5] | Lack of standardized, real-time digital training logs across small-scale mines and contractor pools in Jharkhand. [PS] | Central Web Compliance Dashboard displaying real-time worker certification coverage, module pass rates, and immutable audit logs. [INFERENCE] |
| Dept of Higher & Technical Education (Jharkhand) | Execute SIH mandate; deploy scalable digital vocational training infrastructure across Polytechnic and ITI institutes. [PS] | Existing VR simulator labs are capital-intensive ($\ge \text{₹5-10 Lakhs/unit}$) and non-portable to rural field sites. [PS] | Software-only mobile AR solution running on commodity Android phones ($\approx \text{₹8,000–12,000}$) with public code repo. [INFERENCE] |
| Feature / Module | Target Requirement | Description & Technical Justification |
|---|---|---|
| AR Core Engine (Sceneview / ARCore) | Requirement 1 [PS] | Implements plane tracking, raycasting, and 3D GLB model placement on Android 10+ without external VR headsets or dedicated depth sensors. |
| Fire & Explosion AR Module | Requirement 2 [PS] | Places virtual fire, PASS fire extinguisher (Pull, Aim, Squeeze, Sweep), and directional green exit arrows directly onto camera view of real room. |
| Gas Leak & PPE AR Module | Requirement 2 [PS] | Simulates color-coded toxic gas plume (Methane/$\text{H}_2\text{S}$), forces 3D PPE selection (Self-Rescuer apparatus, safety goggles), and buddy evacuation protocol. |
| Ol Chiki & Devanagari UI Engine | Requirement 3 [PS] | Bundles Noto Sans Ol Chiki TrueType Font and pre-recorded Santali audio assets for complete offline native rendering. [RESEARCHED] [4.2] |
| Offline Room DB & Sync Queue | Requirement 4 [PS] | Stores session data, quiz answers, and signed certificates locally using SQLite/Room; syncs asynchronously when online. |
| Practical & Quiz Assessment Engine | Requirement 5 [PS] | Evaluates visual hazard identification speed, sequence correctness, and multiple-choice visual audio quiz scores. |
| Cryptographic QR Generator & Validator | Requirement 6 [PS] | Generates HMAC-SHA256 signed QR code containing encrypted worker assessment payload for offline/online verification. |
| React Web Admin Portal | Requirement 7 [PS] | Provides DISH/DGMS auditors and mine administrators visual dashboards, worker pass/fail rates, and CSV audit export. |
To ensure technical and legal compliance during development, the team must address the following target research questions:
ArSceneView (Filament renderer) can process at stable $\ge 30\text{ FPS}$ on a Snapdragon 680 / MediaTek Helio G85 device running Android 10? [INFERENCE]TextView correctly render complex Ol Chiki diacritics (e.g., Mu-Gaahlaa, Phaarkaa) without character clipping when using custom .ttf fonts? [RESEARCHED] [4.9]| Data Type | Purpose | Ideal Real-World Source | Hackathon Substitute | Access Method | Availability | Privacy Sensitivity | Confidence |
|---|---|---|---|---|---|---|---|
| Worker Identity Data | Match worker to certificate and company roster | Jharkhand State Skill Development Mission (JSDM) / Coal India Portal | Local Mock JSON Dataset (workers.json) |
Bundled App Asset / Local Form | Public/Enterprise | High (Aadhaar / Employee ID) | [ASSUMPTION] |
| 3D Safety Equipment Models | Render AR extinguishers, gas detectors, PPE, machinery | Standard Industrial CAD / Open Asset Libraries | Customized Low-Poly GLB Assets (Blender) | Bundled APK Asset (/assets/models/) |
Fully Available | Low | [KNOWN] |
| Santali Audio Voiceovers | Provide audio guidance for low-literacy workers | Native Santhali Voice Artists / State Cultural Board | Custom AI Voice Synthesis + Human Validation | Pre-rendered MP3s (/assets/audio/sat/) |
Generated / Open | Low | [RESEARCHED] [3.8] |
| Statutory Safety Rules & Guidelines | Assessment question bank & protocol steps | DGMS Safety Circulars & Mines Act 1952 manuals | Structuring DGMS Guidelines into JSON Quiz Bank | Local SQLite DB (safety_rules.db) |
Public Government Domain | Low | [RESEARCHED] [3.2] |
| Completed Assessment Logs | Compliance tracking and dashboard analytics | Central DGMS Compliance Server | Simulated Sync REST API + Local SQLite | HTTP POST when online / Room DB offline | Enterprise Private | Medium | [INFERENCE] |
| Existing Solution | Description | Strengths | Weaknesses | Relevance | Remaining Gap / SurakshaAR Advantage | Source / Confidence |
|---|---|---|---|---|---|---|
| Static Vocational Manuals & Lectures | Traditional 3-day classroom orientation with paper handouts & flipcharts. | Zero technology cost; easy to organize in remote ITIs. | Retention rates below $20%$ after 7 days [PS]; paper certificates easily forged; passive learning. [PS] | Current baseline used in most small mines across Jharkhand. | SurakshaAR provides interactive 3D spatial learning, increasing retention by $>3\times$ while making certificate fraud impossible via signed QRs. | Problem Statement & DGMS Reports [PS] [RESEARCHED] [1.5] |
| VR Headset Simulators (e.g., CIL VR Labs) | Dedicated VR headsets (Oculus/Vive) in high-tech simulation centers. | Highly immersive; zero physical hazard during training. | Hardware cost high ($\ge \text{₹1-2 Lakhs/unit}$); requires dedicated power/PC; non-portable; causes motion sickness in non-tech workers. [PS] | Used selectively by Tier-1 companies (Tata Steel, Coal India HQ). | SurakshaAR runs on existing worker smartphones ($\text{₹0}$ additional hardware cost), completely portable to field mine sites. | Field Research & Industry Standard [KNOWN] |
| Generic Web-based Safety LMS (e.g., Swayam / iGOT) | Video-based online portal with multiple-choice quizzes. | Easy to track completion; scalable server architecture. | Requires continuous broadband internet (fails underground/rural) [PS]; no physical spatial interaction; English/Hindi only. | Digital safety initiative baseline in India. | SurakshaAR works $100%$ offline, supports native Santali (Ol Chiki), and uses AR spatial interactions instead of simple video watching. | Government Portals [KNOWN] |
+-----------------------------------------------------------------------------------------------------------------------+
| CANDIDATE SOLUTION APPROACH COMPARISON |
+------------------------------------+------------------------------------+---------------------------------------------+
| Feature / Dimension | Approach 1: Unity AR Foundation | Approach 2: Native Android Kotlin + |
| | (C# Engine) | Sceneview / ARCore (Filament) |
+------------------------------------+------------------------------------+---------------------------------------------+
| Platform Architecture | Unity Cross-Platform Monolith | Native Android APK (Jetpack Compose + |
| | | Kotlin) |
| APK Size | Heavy (~150 MB - 250 MB) | Extremely Lightweight (~35 MB - 50 MB) |
| Low-RAM Performance (2GB-4GB) | Moderate (High background overhead)| Superior (Native Filament C++ rendering engine)|
| Native Font Support (Ol Chiki) | Complex (Requires custom TextMesh | Seamless (Native Android `Typeface` & |
| | Pro shader setup) | `Paint` engine) |
| Offline DB & Crypto Integration | Requires third-party Unity plugins | Native Jetpack Room + Java Security Cryptography|
| Development Velocity for Students | Fast for game designers | Fast for Android developers |
| Hackathon MVP Feasibility | High (Visually rich) | Exceptional (Smaller footprint, faster load)|
| Production Deployment Feasibility | Moderate (Hard to bundle with state| High (Standard enterprise Android deployment)|
| | government LMS apps) | |
+------------------------------------+------------------------------------+---------------------------------------------+
Sceneview (Filament-based Kotlin successor to Sceneform) for 3D/AR rendering. [RESEARCHED] [1.2]Noto Sans Ol Chiki), seamless integration with Jetpack Room DB, Android WorkManager, and native Java Cryptography APIs (java.security).We select Approach 2: Native Android Kotlin with Sceneview (Filament + ARCore) for the mobile application, paired with a React/Node.js Web Admin Dashboard for cloud compliance management.
+----------------------------------------------------+
| NATIVE KOTLIN MOBILE CLIENT (APPROACH 2) |
| |
| +--------------------+ +--------------------+ |
| | Jetpack Compose | | Sceneview SDK | |
| | (UI / Ol Chiki TTF)| | (Filament / ARCore)| |
| +---------+----------+ +---------+----------+ |
| | | |
| +------------+------------+ |
| | |
| +-----------v-----------+ |
| | Local Jetpack Room | |
| | (SQLite DB) | |
| +-----------+-----------+ |
+-------------------------|--------------------------+
|
(Async Network Sync)
|
v
+----------------------------------------------------+
| CLOUD ADMIN COMPLIANCE PORTAL |
| |
| +--------------------+ +--------------------+ |
| | Node.js / Express| | React Dashboard | |
| | REST API | | (Auditor Portal) | |
| +--------------------+ +--------------------+ |
+----------------------------------------------------+
+-----------------------------------------------------------------------------------------------------------------------+
| SURAKSHA-AR SYSTEM ARCHITECTURE |
+-----------------------------------------------------------------------------------------------------------------------+
[ LAYER 1: USER INTERACTION & LOCALISATION LAYER ]
+-------------------------------------------------------------------------------------------------------------------+
| - Jetpack Compose Declarative UI Engine - Custom TrueType Font Renderer (Noto Sans Ol Chiki) |
| - Bilingual Audio Engine (Hindi / Santali MP3s) - High-Contrast Iconographic Navigation Engine |
+-------------------------------------------------------------------------------------------------------------------+
|
v
[ LAYER 2: AR SPATIAL SIMULATION ENGINE (SCENEVIEW / FILAMENT) ]
+-------------------------------------------------------------------------------------------------------------------+
| - ARCore Motion Tracking & Plane Detection - PBR 3D Asset Loader (GLTF / GLB Parser) |
| - Raycasting & Collision Detection Engine - Spatial Gesture Recognizer (Tap, Sweep, Drag) |
+-------------------------------------------------------------------------------------------------------------------+
|
v
[ LAYER 3: ASSESSMENT & CRYPTOGRAPHIC CERTIFICATION ENGINE ]
+-------------------------------------------------------------------------------------------------------------------+
| - Scoring & Evaluation Rules Engine - Local HMAC-SHA256 Signer / Keypair Manager |
| - Dynamic Vector QR Code Generator - PDF Certificate Exporter (Android Canvas) |
+-------------------------------------------------------------------------------------------------------------------+
|
v
[ LAYER 4: OFFLINE STORAGE & DATA PERSISTENCE ]
+-------------------------------------------------------------------------------------------------------------------+
| - Jetpack Room SQLite Database - Encrypted Shared Preferences (Credentials) |
| - Android WorkManager (Deferred Queue Sync) - Local Media & 3D Asset Cache (/assets/raw/) |
+-------------------------------------------------------------------------------------------------------------------+
|
(REST API Sync via HTTPS when online)
|
v
[ LAYER 5: CLOUD ADMIN & COMPLIANCE PORTAL (WEB) ]
+-------------------------------------------------------------------------------------------------------------------+
| - Node.js / Express REST API Server - PostgreSQL / MongoDB Central Ledger |
| - React.js Admin Compliance Dashboard - DGMS & DISH Audit Report Exporter (CSV/PDF) |
+-------------------------------------------------------------------------------------------------------------------+
ArSceneView initializes Google ARCore, asking the worker to scan the surrounding floor to detect horizontal/vertical surfaces.WorkerID|ModuleID|Score|Timestamp|DeviceUUID. The Java Cryptography library signs this string using HMAC-SHA256 with a pre-shared app key, rendering it into a high-density QR code rendered on a digital certificate.WorkManager silently triggers an HTTP POST request to sync logs with the Web Admin Portal.| Layer | Selected Technology | Why Chosen | Alternative Considered | Why Alternative Rejected |
|---|---|---|---|---|
| Mobile OS Target | Android 10+ (API 29–34) | Covers $>95%$ of active smartphones used by industrial workers in Jharkhand. [KNOWN] | Android 13+ only | Excludes low-income contract workers with older devices. |
| Mobile UI Framework | Native Android (Kotlin + Jetpack Compose) | Native performance, instant startup, flawless custom font support (Noto Sans Ol Chiki). [RESEARCHED] [4.2] |
Flutter | Platform view wrappers for ARCore in Flutter are unstable and add extra latency. |
| 3D / AR Engine | Sceneview SDK v2.x (Filament + ARCore) | Kotlin-native, light footprint ($\approx 10\text{MB}$ library size), uses Google's Filament PBR renderer. [RESEARCHED] [1.1] | Unity AR Foundation | Adds $>150\text{MB}$ overhead; heavy CPU/GPU load on budget devices; slow cold boot. |
| Local Database | Jetpack Room (SQLite) | Standard Android ORM, supports transactional integrity, reactive Flow streams, zero network dependency. | Realm / Firebase | Firebase requires network connectivity; Realm increases binary size. |
| Background Sync | Android WorkManager | Guarantees deferred execution of network sync jobs even if app is closed or phone reboots. | AlarmManager / Custom Service | WorkManager handles network constraint checks (CONNECTED) automatically. |
| QR & Crypto Engine | ZXing Embedded + Java HMAC-SHA256 | Lightweight, robust QR generation/scanning without cloud dependencies. | Cloud Verification API | Fails underground without internet access. |
| Backend API | Node.js (Express) or Python (FastAPI) | Lightweight REST endpoints, fast JSON serialization, easy containerization. | Java Spring Boot | Heavy memory footprint for simple compliance CRUD services. |
| Admin Frontend | React.js (TypeScript) + Tailwind CSS | Fast development velocity, rich chart libraries (Recharts), mobile-responsive for field inspectors. | Angular | Steeper learning curve for student hackathon timeline. |
| Cloud Database | PostgreSQL | Relational integrity for worker rosters, audit logs, and compliance records with JSONB support. | MongoDB | PostgreSQL offers better relational guarantees for compliance tracking. |
To maintain strict engineering rigor, AI components are evaluated below against the mandatory 14-point criteria:
Score = (Correct_Actions / Total_Actions) * 100). AI/LLM evaluation must never decide whether a miner passes a statutory safety test due to non-deterministic hallucination risks.-- Worker Profile Table
CREATE TABLE workers (
worker_id TEXT PRIMARY KEY NOT NULL,
full_name TEXT NOT NULL,
native_language TEXT NOT NULL, -- 'HI' or 'SAT'
sector TEXT NOT NULL, -- 'MINING', 'STEEL', 'MICA'
assigned_mine_id TEXT NOT NULL
);
-- Safety Training Modules Table
CREATE TABLE modules (
module_id TEXT PRIMARY KEY NOT NULL,
title_hi TEXT NOT NULL,
title_sat TEXT NOT NULL,
domain_category TEXT NOT NULL, -- 'FIRE', 'GAS', 'MACHINERY', 'ROOF_FALL', 'FIRST_AID'
min_passing_score INTEGER DEFAULT 80
);
-- Assessment Results & Completion Ledger
CREATE TABLE assessment_logs (
log_id TEXT PRIMARY KEY NOT NULL,
worker_id TEXT NOT NULL,
module_id TEXT NOT NULL,
score INTEGER NOT NULL,
passed INTEGER NOT NULL, -- 1 for True, 0 for False
time_taken_seconds INTEGER NOT NULL,
completed_at INTEGER NOT NULL, -- Epoch timestamp
digital_signature TEXT NOT NULL, -- HMAC-SHA256 Hash
is_synced INTEGER DEFAULT 0, -- 0 = Pending Sync, 1 = Synced
FOREIGN KEY(worker_id) REFERENCES workers(worker_id),
FOREIGN KEY(module_id) REFERENCES modules(module_id)
);
assets/ directory. Zero runtime network download is needed for training.is_synced = 0. Android WorkManager executes a periodic background task with NetworkType.CONNECTED constraint.is_synced = 1 on the client via HTTP 200 response.To prevent fraudulent certificate generation (a major issue cited in statutory audit background [PS]):
+-------------------------------------------------------------------------------------------------------------------+
| CRYPTOGRAPHIC QR VERIFICATION WORKFLOW |
+-------------------------------------------------------------------------------------------------------------------+
[ STEP 1: ASSESSMENT PASSED (MOBILE CLIENT) ]
Worker completes module with Score >= 80%
|
v
[ STEP 2: PAYLOAD CONSTRUCTION ]
Payload String = "WID:10042|MID:MOD02|SCORE:90|TS:1725500000"
|
v
[ STEP 3: HMAC-SHA256 SIGNATURE GENERATION ]
Signature = HMAC-SHA256(Payload, App_Secret_Key)
|
v
[ STEP 4: QR CODE ENCODING ]
QR Code Data = Base64(Payload + "." + Signature)
|
v
[ STEP 5: OFFLINE VERIFICATION (INSPECTOR DEVICE) ]
Inspector scans QR code using SurakshaAR Inspector Mode
App splits Payload and Signature -> Recomputes HMAC using same Key
Match = VALID CERTIFICATE (Green Checkmark + Details)
Mismatch = TAMPERED / INVALID (Red Warning)
+-----------------------------------------------------------------------------------------------------------------------+
| HACKATHON MVP BUILD SCOPE |
+----------------------------------------------------+------------------------------------------------------------------+
| IN SCOPE (MUST BUILD FOR SIH DEMO) | DESCRIPED ONLY / OUT OF SCOPE FOR HACKATHON |
+----------------------------------------------------+------------------------------------------------------------------+
| 1. Functional Android APK (Android 10+ support). | 1. Full 5-domain module suite (Only 2 full domains in MVP). |
| 2. AR Module 1: Fire & Explosion Response | 2. Hardware IoT Gas Sensor integration via Bluetooth. |
| (Extinguisher PASS sequence + Exit overlay). | 3. Multi-tenant Enterprise SSO integration (Active Directory). |
| 3. AR Module 2: Gas Leak & PPE Protocol | 4. Cloud AI automated Santali speech recognition (STT). |
| (Hazard zone identification + PPE selection). | 5. Wearable Smart Glass (HoloLens/RealWear) support. |
| 4. Bilingual Localisation: Hindi & Santali | |
| (Ol Chiki font rendering + Audio cues). | |
| 5. Offline Practical Assessment Engine & Scoring. | |
| 6. Cryptographic QR Certificate Generation & Scan. | |
| 7. Web Admin Compliance Dashboard | |
| (Worker list, pass rates, audit logs). | |
| 8. Public GitHub Repo & Demo Video. | |
+----------------------------------------------------+------------------------------------------------------------------+
During the live SIH jury demonstration, judges will evaluate the system in a standard indoor room without access to a real coal mine or live industrial hazard. The team will simulate the field environment cleanly and transparently as follows:
+-----------------------------------------------------------------------------------------------------------------------+
| SIH LIVE DEMO SIMULATION PROTOCOL |
+------------------------------------+------------------------------------+---------------------------------------------+
| Environment / Feed | Real-World Production State | Live SIH Judging Simulation Strategy |
+------------------------------------+------------------------------------+---------------------------------------------+
| Physical Mine Room / Tunnel | Underground coal seam drift or | Standard hackathon presentation booth floor |
| | steel plant floor | scanned via phone camera in real-time. |
| Emergency Fire Hazard | Live industrial chemical / coal | Simulated 3D animated fire asset overlaid |
| | dust fire | on booth desk via ARCore plane detection. |
| Toxic Gas Leak ($\text{CH}_4$) | Hazardous gas buildup detected by | Augmented 3D color-coded particle plume |
| | methanometer | rendered in camera view. |
| Network Connectivity | Zero cellular / underground pit | Phone placed in AIRPLANE MODE during demo |
| | | to prove 100% offline AR & QR generation. |
| Verification Inspection | Gate Inspector scanning worker | Second smartphone used in Airplane Mode to |
| | certificate | scan generated QR and verify authenticity. |
+------------------------------------+------------------------------------+---------------------------------------------+
2026-09-05 2026-09-12 2026-09-20 2026-09-26 2026-09-30
|-----------------------|-----------------------|-----------------------|------------------| (Deadline)
Phase 1: Core Setup Phase 2: AR Modules Phase 3: Crypto & Sync Phase 4: Testing Final Polish
& Localisation & Assessment Engine & Web Admin Portal & Video Demo & Submission
Sceneview SDK v2.x.Noto Sans Ol Chiki TTF font and set up bilingual string resource dictionaries (HI/SAT). [RESEARCHED] [4.2].glb files under $5\text{ MB}$ total.WorkManager sync task for auto-upload when network is restored.| Test Category | Target Objective | Test Method & Tool | Success Criteria |
|---|---|---|---|
| AR Motion Tracking | Stable plane detection on mid-range phones | Test on Android 10 device with 3GB RAM (Redmi Note 9) | Surface detected within $< 3\text{ seconds}$; 3D model drift $< 2\text{ cm}$. |
| Offline Capability | $100%$ functionality without internet | Enable Airplane Mode; execute full module, quiz, and QR generation | Zero crash, zero network error popups, certificate generated successfully. |
| Font Rendering (Ol Chiki) | Accurate text display in Santali | Render 50 complex Ol Chiki character strings in Compose | Zero box glyphs ($\square$), zero character overlapping. [RESEARCHED] [4.9] |
| QR Security & Tampering | Reject altered certificate payloads | Modify Base64 QR payload string by 1 character and scan | Verification engine flags status: "TAMPERED / INVALID CERTIFICATE". |
| App Footprint | Low storage and fast launch | Inspect final release APK build size via APK Analyzer | APK size $\le 45\text{ MB}$; cold boot time $< 3\text{ seconds}$. |
| Web Sync Reliability | Background queue execution | Complete 5 offline modules, connect Wi-Fi, monitor WorkManager | All 5 records successfully POSTed to Web Admin; is_synced updated to 1. |
| Risk Event | Likelihood | Impact | Severity | Mitigation Strategy |
|---|---|---|---|---|
| ARCore Compatibility Failure on Budget Phones | Medium | High | High | Fallback to a 3D Interactive Touch Mode (Filament 3D view without camera tracking) if ARCore initialization fails on non-supported hardware. |
| Font Rendering Clipping for Ol Chiki Script | Low | Medium | Medium | Embed Noto Sans Ol Chiki directly in app assets and set explicit line-height padding in Compose TextStyle. [RESEARCHED] [4.2] |
| High Camera Battery Drain during AR Drills | High | Medium | Medium | Auto-terminate camera feed after module completion; cap AR frame rate to $30\text{ FPS}$; keep training drills concise ($3–5\text{ mins}$). |
| Certificate QR Code Fraud / Key Leakage | Low | High | High | Use HMAC key obfuscation via Android NDK C++ native library (.so) to prevent reverse-engineering of secret key from APK. |
| Parameter | Generic SIH Submissions | SurakshaAR Solution |
|---|---|---|
| AR Engine Choice | Heavy Unity builds ($>200\text{MB}$) that lag on mid-range Android phones. | Native Kotlin + Sceneview (Filament) yielding an ultra-fast $35\text{MB}$ APK. [RESEARCHED] [1.2] |
| Localisation Strategy | Google Translate API calls (fails offline; poor Ol Chiki support). | Embedded Noto Sans Ol Chiki fonts + offline pre-recorded native Santali voiceovers. [RESEARCHED] [4.2] |
| Network Reliance | Requires cloud servers for quiz grading and certificate creation. | $100%$ offline operational capability with local SQLite DB and HMAC-SHA256 QR signing. |
| Verification Fraud | Simple unverified PDF certificates or static QR links. | Cryptographically signed HMAC QR code verifiable offline by inspectors without internet access. |
| Regulatory Alignment | Generic educational content. | Tailored to statutory mandates under Mines Act, 1952, Factories Act, 1948, and DGMS guidelines. [RESEARCHED] [3.2] |
Noto Sans Ol Chiki font provides complete Unicode compliance for Santali script on Android OS. [4.2]Sceneview SDK (Filament renderer) provides lightweight native Kotlin 3D/AR execution on Android without Unity engine overhead. [1.2]io.github.sceneview:arsceneview dependency. [RESEARCHED] [1.7]Noto Sans Ol Chiki font into res/font/ and test Compose UI rendering of Santali strings. [RESEARCHED] [4.2].glb..glb asset polycount is $< 15,000$ vertices per model.Directorate General of Mines Safety (DGMS), Dhanbad
Google Developers — ARCore & Font Documentation
Sceneview Android Repository
Ministry of Labour & Employment / Government of India
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.