A deep learning approach for motion segment estimation for pipe leak detection robot

Water is vital for all living beings, especially for a human. Automatic position detection of water leakage in water pipelines is very important to minimize the loss of labour, time, money spent on exploration and excavation in pipe inspection procedures. The main goal of detection is to prevent water loss. In this study, sensitive position detection, crack frequency band detection and external sphere studies of an in-pipe robot prototype have performed. During the precise position estimation, classical EKF, stationary region detection and location estimation using EHDE are performed with two different ANNs. In this way, online precise position estimation can be done on hardware that has not sufficient computational power for indoor robotic studies. In addition, the sound characteristics resulting from the crack at different hole size and water pressure intensity levels have investigated. Finally, a new sealing sphere design has devised and three different hydrophone sensor data have recorded on the SD card simultaneously. It has been found that the proposed ANN method has the performance to work online and can make a similar position estimation with the classical IMU position estimation method by 99%.

show abstract

“…In Equation (24), The Jacobian 𝑭 𝒕 transition matrix of the 𝒇(𝒙 𝒕−𝟏 ) function obtained when Equation ( 23) is replaced in Equation (21).…”

Section: Integration Stepmentioning

confidence: 99%

Position Estimation of In-Pipe Robot using Artificial Neural Network and Sensor Fusion

Akkaya

Talu

Aydoğmuş

2021

Sakarya University Journal of Science

View full text Add to dashboard Cite

show abstract

“…Pipe inspection robot is developed and investigated in [16]. Pipe leak detection robot is analysed in [17]. Method of design of a pipe robot is described in [18].…”

Section: Introductionmentioning

confidence: 99%

Dynamics of a two mass pipe robot with the self-stopping mechanism based on viscous friction

Ragulskis¹,

Spruogis²,

Bogdevičius³

et al. 2020

Math. models, eng.

View full text Add to dashboard Cite

A model having two degrees of freedom and incorporating a specific type of nonlinearity is proposed for the analysis of dynamics of a pipe robot. Specific type of nonlinearity has different values of viscous friction depending on the sign of velocity of the system. Numerical investigations for various parameters of the system are presented. Optimal frequency of excitation of a pipe robot is determined. The obtained results are used in the process of design of pipe robots: it is especially important to choose the frequency of excitation in order to ensure effective operation of a pipe robot.

show abstract

Three-dimensional mapping of pipelines using laser ranging and a gyroscope

Cui,

Liang

2023

Sci Rep

View full text Add to dashboard Cite

Conventional underground gas pipeline equipment can only map straight pipelines at close distances. As such, a dual-robot mapping technique is proposed in this study, based on laser ranging and a gyroscope, for application to straight, bent, or sloped underground pipelines. This process is facilitated by two robots (a detecting vehicle and a mapping vehicle) alternately moving from the entrance to the exit of a pipe system. The two vehicles acquire data using individual distance sensors and a gyroscope, producing 3D maps of each section. Mapping results can then be produced for the entire pipeline by splicing images from multiple sections. Simulated validation experiments were conducted using horizontal straight, bent horizontal fixed turning radius, sloped uphill or downhill, and composite sections. Results showed this approach could effectively map different types of pipelines with inner diameters of 300–500 mm. Distance errors over 2000 m of travel were within 1 m, while the angular error was within 1.5°, demonstrating our approach to be highly accurate for complex pipeline system mapping.

show abstract

A deep learning approach for motion segment estimation for pipe leak detection robot

Cited by 3 publications

References 8 publications

Position Estimation of In-Pipe Robot using Artificial Neural Network and Sensor Fusion

Position Estimation of In-Pipe Robot using Artificial Neural Network and Sensor Fusion

Dynamics of a two mass pipe robot with the self-stopping mechanism based on viscous friction

Three-dimensional mapping of pipelines using laser ranging and a gyroscope

Contact Info

Product

Resources

About