Bengaluru Metro to use AI, drones for monitoring concrete structures

Representational picture

Bengaluru Metro Rail Corporation Ltd (BMRCL) plans to use artificial intelligence (AI) and drones to monitor the safety of concrete structures, including viaducts and pillars.

BMRCL director (operations and maintenance) Sumit Bhatnagar told Moneycontrol that drones equipped with high-resolution camera tools will collect data on structural conditions. AI tools will analyse the data to detect issues such as cracks, degradation or honeycomb patterns in the concrete.

BMRCL is planning to invite a tender for the AI-drone based inspection of Reach 1 (MG Road – Byappanahalli), the first operational Metro corridor in Bengaluru, inaugurated in 2011.

Also read: AI Cops! Bengaluru police pilots AI-generated avatars for social media outreach

Specifications for the system are being finalised before the tender is floated, Bhatnagar said. “AI-based systems will quantify the severity of damages and issue warnings which enables our engineers to take timely corrective actions,” he said.

Monitoring will focus on viaducts and pillars, which have shown instances of cracks and honeycomb formations.

“AI-powered analysis will identify whether defects are harmful and recommend corrective actions such as sealing with epoxy or other interventions,” Bhatnagar said.

Defects often arise during the construction phase due to undetected flaws. Though such issues are infrequent, addressing serious cases promptly is critical for ensuring safety of structures. “Currently, structural inspections are carried out manually using binoculars, cameras, and hydraulic platforms. The adoption of drones and AI will reduce manual effort, improve accuracy, and allow for faster assessments of the condition of the infrastructure,” Bhatnagar said.

Drones with sensors will be used to collect data will reduce the time, risks, and costs associated with manual inspections.

"For instance, traffic was stopped or redirected during such inspections as they required the use of hydraulic platforms." he said.

The data will be processed using AI-based deep learning algorithms and 3D models to detect and classify defects automatically. The system will quantify the severity of each defect in terms of size, type, and location which enables data-driven decisions for maintenance.

The processed data will be visualised on a cloud-based asset management platform, providing detailed summaries and tools for structural health assessment.

BMRCL is also planning to use AI to monitor Yellow Line, which connects RV Road to Bommasandra and is expected to become operational by Q2 2025. This line will link key locations such as Electronics City, home to companies such as Infosys and Biocon.

Also read: In a first, Bengaluru Metro will use AI to monitor tracks

On the Yellow Line, a pilot train equipped with a track-monitoring system will check the condition of tracks. Images of the tracks will be transmitted to a central server, where AI will analyse them and generate alerts for anomalies.

In cases of severe problems, the system will send immediate alerts to the Operation Control Centre through the Train Control and Management System (TCMS). To ensure reliability, the monitoring system will be installed on two train sets, providing a backup in case of defects in one set.