Towards an Automated and Auditable Industrial Safety Inspection Using Robots and Blockchain
摘要
This study addresses the challenges of ensuring data integrity, transparency, and security in industrial safety inspections, particularly in hazardous environments like the mining industry. The research investigates how a blockchain-based system can improve data management and trust in robot-assisted industrial inspections.
MethodsA novel blockchain-based inspection management system is proposed, integrating blockchain technology with the Robot Operating System (ROS) to enhance the integrity and traceability of inspection data. The system automates safety inspections through ROS-compatible smart contracts and uses the InterPlanetary File System (IPFS) for scalable, decentralized storage of large files such as ROS bags. The evaluation was conducted using realistic data collected by EspeleoRobô, a semi-autonomous robot, during industrial inspections in a mining environment.A novel blockchain-based inspection management system is proposed, integrating blockchain technology with the Robot Operating System (ROS) to enhance the integrity and traceability of inspection data. The system automates safety inspections through ROS-compatible smart contracts and uses the InterPlanetary File System (IPFS) for scalable, decentralized storage of large files such as ROS bags. The evaluation was conducted using realistic data collected by EspeleoRobô, a semi-autonomous robot, during industrial inspections in a mining environment.
ResultsThe implementation demonstrated improved security, transparency, and traceability in industrial inspection processes. The blockchain-based approach ensured that inspection records remained immutable and verifiable. Smart contract deployment sizes ranged from 1.6 to 12.9 kB, with system scalability evaluated based on the number of inspection locations, standards, and components. ROS message sizes and transaction times were analyzed, with odometry messages averaging 914 bytes and transaction times around 2 s.
ConclusionThe study concludes that integrating blockchain, ROS, and IPFS provides a viable solution for enhancing the security, transparency, and scalability of industrial inspection processes using robots. This approach offers improvements over traditional data management systems, making it promising for industries with complex, data-intensive environments, such as mining. The system also addresses data integrity during network communication loss through local storage with Merkle tree verification.