Purushothaman Balamuralidhar

Image Analysis-Based Automatic Detection of Transmission Towers using Aerial Imagery

Uninterrupted electricity transmission is a critical utility service for any nation. A major component of nation-wide infrastructure carrying electricity are the transmission towers. To give uninterrupted supply, timely maintenance of towers is a must. Due to vastness of power grid, fault detection via aerial inspection and imaging is emerging as a popular method. In this paper, we attend to the problem of automatic detection of towers in specific images. We present a four-stage algorithm for such detection. For a porous, cage like object structure that of a tower, we use gradient density and a novel feature called cluster density to detect pylon blocks. The algorithm was tested against image data captured for many towers along two different power grid corridors. The algorithm demonstrated missed detection of (<) 1 % and complete absence of false positives, which is very encouraging. We believe that our result is far more useful in tower detection, than available previous works.

Power infrastructure monitoring and damage detection using drone captured images

Infrastructure detection and monitoring is a difficult task. Due to the advances in unmanned vehicles and image analytics, it is possible to decrease the human effort and achieve consistent results in infrastructure assessments using aerial image processing. Reliable detection and integrity checking of power infrastructure including conductor lines, pylons and insulators in a diverse background is the most challenging task in drone based automatic infrastructure monitoring. Most techniques in literature use first principle approach that tries to represent the image as features of interest. This paper proposes a deep learning approach for power infrastructure detection. Graph based post processing is applied for improving the outcomes of the generated deep model. A f-score of 75% is achieved using the deep model which is further improved using spectral clustering for the conductor lines, pylons and insulators that form the core parts of power infrastructure.

An efficient backtracking-based approach to turn-constrained path planning for aerial mobile robots

Many important classes of civilian applications of Unmanned Aerial Vehicles, such as the class of remote monitoring of long linear infrastructures e.g. power grid, gas pipeline etc. entail usage of fixed-wing vehicles. Such vehicles are known to be constrained with restricted angular movement. Similarly, mobile robots such as car robots or tractor-trailer robots are also known to entail such constraint. The algorithms known so far require a lot of preprocessing for turn constraint. In this paper, we introduce a novel algorithm for turn angle- constrained path planning. The proposed algorithm uses a greedy backtracking strategy to satisfy the constraint, which minimizes the amount of backtracking involved. By further constructing an efficient depth-first brute-force algorithm for path planning and comparing against its performance, we see an improvement in convergence performance by a factor of at least 10x. Further, compared to recent LIAN suite of path-planning algorithm, our algorithm exhibits much reduced discretization offset/error with respect to shortest path length. We believe that this algorithm will form an useful stepping stone towards evolution of better path planning algorithm for specific mobile robots such as UAVs.

A Novel Cognitive Cycle for Fault Diagnosis in Infrastructural Systems

Abstract Remote sensing techniques are being increasingly used for periodic structural health monitoring of vast infrastructures such as power transmission systems. The current efforts concentrate on analysis of visual and other signals captured from the sensing devices, to diagnose the faults. Such data collection and analysis is expensive in terms of both computational overheads as well as towards robotic maneuvering of the data collection platform, such as a UAV. In this paper, we model the data gathering platform as an intelligent situated agent, and propose to autonomously control its data gathering and analysis activities through a cognitive cycle, to optimize the cost of efforts in identifying the faults that may exist. In this context, we explore use of less expensive qualitative reasoning with the background knowledge expressed as a Qualitative Bayesian Network (QBN). We introduce a reactive, economical planning algorithm around QBN that controls the sequence of data collection and analysis, much like how human inspectors do. We substantiate our claims with the results of simulation of the corresponding cognitive cycle.