Pipelines are the critical infrastructure systems in US. All these pipeline networks serve our daily life by providing sewers, storm and main water. However, all of these networks have their limited service life and, thus, it is critically needed to assess their conditions periodically throughout their life cycle before any accidents occur due to defects and deterioration. For example, leakage from a sewage or water line triggers soil erosion and impacts the environment. Soil erosion causes a road to collapse immediately. Such incidents seriously disturb our daily life such as cutting off services provided by the utilities network. In addition, pipeline failure can result in serious ecological disasters and human casualties as well as financial loss. Therefore, here we propose to develop a mapping and nondestructive test system using advanced technologies for condition assessment of pipeline networks. Several pipeline leak detection methods have been developed and used during the last decades. The exterior approach utilizes various human-made sensing systems to achieve the detection task outside pipelines. Moreover, the biological approach utilizes visual, auditory and/or olfactory senses of trained dogs or experienced personnel to detect leakage. Methods involving animals are not effective for prolonged operation for more than 30–120 min of continuous searching due to fatigue.
It has been shown that remotely operated vehicles are durable for performing pipeline inspection tasks and functioning where cannot be accessible by dog, pigging or human. In the past decades, many technologies have been developed and used for condition assessment of pipeline networks: closed-circuit television (CCTV), acoustic methods, and lately laser-based scanning and GPR. Beyond the above techniques, infrared thermography (IRT) is one of the possible methods for pipeline condition assessment that has attracted less attention. One of the major limitations of IRT is its inability to work under water, so it can only study liner delamination in the pipe crown where the surface is exposed to the air. A tailored covariance function for pipeline cylindrical structures within the context of a Gaussian Processes has also been developed to regress missing sensor data incurred by a sampling strategy followed in the field to speed up the inspection times, given the slow response of the pulsed eddy current electromagnetic sensor proposed. The project will focus on the inspection of open-end culverts and storm drain systems. Thus, the main materials to be investigated are reinforced concrete and corrugated metal. Other types of materials can also be considered including poly propylene and polyester. For each material, the sensors should be selected, and the calibration must be conducted. The AI-based algorithms should be tuned accordingly.
The development of autonomous vehicles in pipeline inspection and monitoring, which is proposed here, will reduce the extent of human operator involvement in unmanned vehicles through the implementation of intelligent control algorithms. More safe and accurate remote monitoring can be achieved by using smart robots and sensor networks equipped with advanced artificial intelligence-based techniques. In the systems, ground robots/crawlers will be used with onboard sensing enable environmental observation. The ground robots will be equipped to enable remote communication, teleoperation, and the transmission of collected data. Effective automated mapping requires the flexibility to allow real-time changes in planning to accommodate unmodeled/unexpected uncertainties and disturbances. Ground robots shown in the Figures 1-2 below will play essential parts in the task of mapping and non-destructive testing (NDT) of hydraulic pipelines, where human access is highly restricted. Once the autonomous system proposed here is developed, it can be used over and over with reduced cost only for maintenance. In addition, making the process autonomous would reduce the risk by human inspection, and save time for inspection, and less trained persons can obtain the data.