Knowing where things are underground has become important enough that in a number of jurisdictions around the world including the Netherlands, Singapore, and the U.K., initiatives are underway to create national digital twins which include below-ground infrastructure. Recently there have been important technical advances that are have brought the feasibility of cost-effective mapping of underground infrastructure within reach. The most accurate way to detect the location of underground infrastructure is to (carefully) dig a hole to expose it and then bring in a survey team to survey the location. But this is time consuming and expensive. It has been estimated that the average direct and indirect cost of a pothole is $30,000. Depending on the technology used to dig the hole it can also he hazardous. Current best practice for remote locating underground infrastructure is walking the site with electromagnetic wands (EMI) or ground penetrating radar (GPR) pushcarts. This is slow, expensive, and hazardous.
Most construction projects are only concerned with the underground utility and other facilities that lie within a few meters of the surface. But there is another class of subterranean objects that are large and lie deeper. These include rail, large scale water transmission, and large diameter storm, sanitary and combined sewer systems. The traditional approach to investigating such facilities is CCTV. These are TV cameras connected to a cable for power and image transmission that are guided down pipes. While such systems are low cost, they don’t capture location data, they have limited range requiring multiple deployments from manholes, and assessment and reporting is manual and arduous.
I have blogged previously about inertial locating and mapping, which relies of accelerometers, gyroscopes, on devices with memory for recording output from the sensors and a battery to power the device. Such devices do not require a cable connection for power and data transmission. Inertial mapping technology is composed of an Orientation Measurement Unit (OMU) that contains inertial sensors. The OMU is battery powered and the data logged is stored internally during a measurement run. Unlike other technologies for tracking underground facilities the device does not need to be traced from above ground as it progresses through a pipe. The device also contains odometers recording the speed of travel. After a run post-processing software converts the data recorded by the OMU and odometers to an accurate 3D model.
Now a recent a startup in the UK, Headlight AI, has developed a system that builds on inertial technology but adds significant capabilities for mapping and condition assessment of large deep assets. I had a chance to chat with Dr. Puneet Chhabra, Co-founder and CTO at Headlight AI. Headlight AI’s product Telesto integrates inertial technology with wireless communication, LiDAR, cameras, and lights to capture location and high precision 3D point cloud data that allows assets to be mapped and high-resolution LiDAR and imagery to be captures at high frequencies as the device progresses through an underground asset. Wireless communication enables real-time monitoring of the progress of the device at shorter ranges. In addition, a machine learning application has been developed which helps automate the detection of different materials and identifying and locating defects such as corrosion, calcification, wall deformations, and other common problems.
An example where the technology has been deployed is the Bournemouth Coastal Interceptor Sewer (BCIS) project, which is 8 km 2.2 m diameter sewer deep underground with flowing water in some sections. The BCIS is a critical asset managed by Wessex Water. Its concrete structure was constructed between 1964 and 1971 with an internal diameter of 1,800mm. Due to its location and its surrounding geology it is particularly vulnerable to water infiltration and biogenic corrosion from hydrogen sulfide.
The BCIS is regularly surveyed by Wessex Water to ensure its effective operation. Surveying methods have included CCTV condition assessments, laser profiling surveys, and calibrated LiDAR surveys. However, this traditional surveying method requires person entry and traversing of the sewer by a team of operatives due to the need for calibration targets and accurate registration of the 3D point cloud. Hence, this method requires substantial assessment and planning in order to accommodate the higher health and safety risks imposed on workers.
Wessex Water’s priority in the management of its sewers is health and safety. In many cases access shafts into sewers can be located every 1000m or so, which is concerning if operatives are to enter such dangerous environments. Therefore, person entry into the confined spaces of a live sewer, such as the BCIS, are minimized or avoided wherever possible to ensure enhanced process safety.
Using Headlight AI’s technology required 2 weeks to collect the data across 7,692 m of the BCIS, including 13 shaft scans at various positions and depths along the sewer. For the sewer alone 250 GB of data and over 2 billion points were collected. Headlight AI’s software made it possible to automatically distinguish different wall materials in the point cloud using a combination of traditional and deep learning algorithms. Furthermore, machine learning helped to identify and locate common defects for follow-up assessment.
The project has enabled approximately 8 km of the Bournemouth CIS to be mapped in 3D without the need of operatives to traverse the sewer. The cost of the Telesto survey was similar in cost to laser profiling and CCTV surveys, while being significantly more informative and less hazardous to workers health and safety. This project has demonstrated that Telesto has the potential to improve the quality and the health and safety of sewer inspections across the water and wastewater industry. In a very recent development it is now it possible by combining Telesto information with terrestrial surveys and shaft scans to map the key sections of deep sewers in x, y, and z. Further future benefits will be the ability to compare how the 3D structure evolves over time and to conduct change detection. This will enhance predictive maintenance and allow water utilities to make more informed decisions regarding both operational and capital expenditure.