Problem Statement

Underground sewer and drainage pipelines are difficult and unsafe to inspect manually. Cracks, blockages, leaks, and structural damage are often detected only after causing flooding, road damage, or system failure.

Solution

Paarvai is an autonomous/semi-autonomous sewer inspection robot using cameras, sensors, IMU and AI vision to detect cracks, blockages and structural defects. It estimates defect locations and records findings on a digital dashboard for maintenance.

In Detail

Paarvai is a compact inspection robot designed to travel through sewer and drainage pipelines from an accessible entry point. Since these pipelines are usually dark, the robot uses high-brightness LEDs to illuminate its path while an onboard camera continuously captures the internal condition of the pipe. The robot can be operated manually or in a semi-autonomous mode. Front-mounted sensors help detect obstacles and automatically stop the robot when something is too close, reducing the risk of collision or damage. The captured video is analyzed using an AI-based defect detection system that can identify conditions such as cracks, blockages, foreign objects, root intrusion, surface damage, and normal pipe sections. Wheel encoders track the distance travelled, while an IMU monitors the robot's orientation. Since GPS signals are generally unavailable underground, this combination helps Paarvai estimate the location of detected defects relative to the pipe entrance. For example, a detected issue can be recorded as: Defect: Crack Location: 3.5 m from entry Severity: Medium Confidence: 92% The detected defects are displayed on a dashboard along with the robot's status, travelled distance, captured images, pipe health condition, and maintenance priority. Once the inspection is completed, Paarvai generates a digital report containing the location and severity of the identified defects. This helps maintenance teams locate problems early and focus repairs on the affected sections instead of manually inspecting the entire pipeline. Future versions of Paarvai can include autonomous junction navigation, 3D pipeline mapping, digital-twin integration, multiple collaborating robots, and automated cleaning or repair mechanisms.