Main content starts here.

Maintaining a sound lifeline and aiming to actualize a smart society with next-generation infrastructure inspection technology

April 17, 2017

Using autonomous flying drones to conduct hammering tests

Water pipes are not the only things that are difficult to access manually. Damage to concrete components and other internal degradation in roads, bridges and tunnels is also difficult to detect. Manual hammering tests are effective means of investigating internal degradation of structures, but they require expensive and time-consuming construction of scaffolding and restriction of traffic.

Considering such circumstances, we have been moving forward with research and development of a solution that involves using small, unmanned multicopters (drone) to carry out inspections at high elevations such as bridges. We call these devices the “flying hammering test robot.” Research for this technology is being conducted by Council for Science, Technology and Innovation, “Cross-ministerial Strategic Innovation Promotion Program (SIP), Infrastructure Maintenance, Renovation, and Management”. (funding agency: NEDO).

The drone under development uses lasers to measure the surroundings and automatically estimates distance and detects faults. The device does not rely on GPS information and can fly autonomously inside tunnels and behind bridge structures where GPS cannot be used.

The drone is equipped with a hammer testing sensor and visual inspection camera. They fly autonomously in close proximity to the structure being inspected and perform the hammering test automatically. Noise-canceling technology is used for accurate isolation of the signal sound. The acoustic and image data can be checked by inspectors by using a terminal on the ground.

We are currently working hard to achieve a practical system, targeting deployment by 2019. Implementation of the system would reduce the need for dangerous work to be performed by people and also facilitate the transfer of veteran knowledge and expertise.

NEC has also developed the world’s first technology for estimating the state of degradation inside a structure from video. The technique combines an “object vibration measurement algorithm” capable of quickly and precisely detecting micro-vibrations in structures based on video images, and “vibration correlation analysis algorithm” capable of identifying and detecting differences in vibration patterns at points where internal deterioration is occurring. This makes it possible to precisely estimate the extent of deterioration inside structures, something that would be impossible by sight alone. By harnessing these technologies, smart inspection can be carried out on roads, bridges, tunnels and other structures.

Focusing on “lines” rather than “points” for early detection of states that are “not usual”

Critical infrastructure such as plants, factories and power generation facilities are also facing new problems besides deterioration due to aging. This is due to the division of physical systems inside facilities into components as a result of advanced specialization. The behavior of complex, large-scale interlinked systems is difficult to understand, even for experts, but invariant analysis is effective for that purpose. This is one technology among NEC’s portfolio of AI technologies which can be used to discover new regularities through data.

Specifically, a very large amount of time series data from a large number of installed sensors is collected and analyzed to automatically extract relationships and model the “usual state”. This means that the values from individual sensors are not understood as “points,” but as “lines,” which are relationships among sensor values. By using this model as the basis for real-time analysis of large amounts of sensor data, a state of “not usual” can be quickly detected.

Because the relationship equations have been simplified so that they can be calculated at high speed, it is possible to make comparisons with a large amount of sensor data that is being collected in real time. This enables discovery of slight changes that even experts cannot easily notice, as well as previously unknown abnormalities, etc. It is therefore possible to detect symptoms in plant facilities or equipment at an early stage and perform preventive maintenance to avert failures before they occur.

In many plants, the common practice is to employ time-based maintenance (TBM), in which inspections are performed periodically as specified by law, etc. However, that approach requires uniform inspection and replacement of parts in facilities and equipment that are sound, which is a large burden. Applying the technology described here, on the other hand, makes it possible to identify trends in degradation and symptoms of failure and otherwise understand the conditions of the plant. That enables condition-based maintenance (CBM) and reliability centered maintenance in which the most suitable maintenance is performed at the optimum time, thus reducing work and cost and contributing to advanced maintenance that ensures a higher level of safety.

The aging of critical infrastructure continues, especially in developed countries. Given limited financial and human resources, maintaining sustainability with a high level of safety is an urgent problem. NEC is combining advanced ICT and knowledge to realize a smart society in which we can all live brighter, more prosperous lives, utilizing the full power of the Group.

Did you enjoy the contents?


Your comments and suggestions are welcome


  • *Please write messages in English. Thank you for your cooperation.
  • *Response to comments will not be made.
  • *All personal information will be handled according to our privacy policy.
Main content ends here.
Top of this page