CarbonX

The system uses AI agents to plan optimal flight paths, collect emission data, and provide detailed environmental analysis, with results securely stored on Starknet. To provide the feasibility of the solution we have a gazebo simulation which simulates an industry environment. At its core, the system deploys AI-powered drones that autonomously monitor industrial carbon emissions using Covalent's Zero-Employee Enterprise (ZEE) framework to orchestrate intelligent agents for path planning and emission analysis. The system's security and data integrity are maintained through Nillion's SecureVault and NilQB, enabling zero-knowledge verification of drone identities and emission data. Each registered drone collects emission data along AI-optimized flight paths, with the data being securely verified

Contracts- Deployed on Starknet: The workflow begins with company registration where the smart contract assigns an initial eco_score and token requirements through register_company(). Companies must acquire and maintain the minimum required tokens before proceeding to device registration. Through register_device(), companies register their drones/sensors, where each device receives a cryptographic key pair essential for data authentication. These registered devices must maintain an active status to participate in the emission reporting process. The operational cycle starts with the AI agent determining optimal flight paths, executed through update_flight_path(). As drones collect data along these paths, they sign the emission data with their private keys before submission through submit_report(). The contract verifies the device signature, updates the emission data in CompanyRegistry, and automatically adjusts token requirements - companies with higher emissions must acquire more tokens, while those with lower emissions can sell their excess. This creates a dynamic token economy where companies can trade tokens through transfer(), while the AI continuously updates eco_scores based on compliance and emission patterns. The entire process ensures transparent, verifiable emission monitoring while incentivizing emission reduction through the token mechanism.

Nillion: The system leverages Nillion's SecureVault and NilQB to create a zero-trust verification environment for drone agents and emission data. When a company registers a device, the drone's unique identifier and cryptographic credentials are stored in SecureVault, with the company's ownership proof distributed across Nillion's network using secure multiparty computation. This architecture ensures that when drones submit emission reports, the smart contract can verify the device's registration and data integrity through SecureVault lookups, while maintaining data privacy and preventing unauthorized access or tampering. The continuous stream of verifiable data from autonomous drone agents is secured through Nillion's network, enabling trustless validation of emission measurements

Covalent: The carbon emission monitoring system leverages Covalent's Zero-Employee Enterprise (ZEE) framework to orchestrate autonomous AI agents in a coordinated workflow. Two primary ZEE workflows drive the system: the Path Planning workflow and the Emission Analysis workflow. The Path Planning workflow combines a Weather Analysis agent and Path Planning agent to determine optimal monitoring coordinates based on weather conditions and geographical constraints. Each agent utilizes specialized tools through ZEE's tool integration system - the Weather agent processes environmental data while the Path Planning agent calculates strategic monitoring points. The Emission Analysis workflow then takes these coordinates and employs an Emission Analysis agent to process sensor data and generate standardized emission scores on a 1-100 scale. These workflows maintain state through ZEE's state management system, allowing for continuous monitoring and updates as conditions change. Through ZEE's orchestration, the agents work in concert to maintain an automated, intelligent monitoring system that requires minimal human intervention while ensuring accurate and consistent emission data collection and analysis.