Smart Cities-Automatic Power Lines Inspection

Abstract

This paper describes an ongoing development of an Unmanned Aircraft System (UAS), with embedded Artificial Intelligence (A.I.) for visual inspection of power lines over urban regions. The development considers the concept of Drone as a Service Tool, together with the Smart Cities, relating all equipment as an Internet of Things (IoT) device, bringing to the urban scenario the aerial mobility and the A.I procedures, to increase the process effectiveness and reducing costs of the orthodoxy power lines inspection process, with the use of new technologies. The application development was made considering the operational requests of the Brazilian National Agency of Electric Energy, that is, considering the automatization activity at the maintenance process on the field an improvement. In this case, the Aerial platform, is represented by a quadcopter with embedded computer running A.I. procedures and computer vision, to control and to do the visual inspection, according to the mission requirements. The integration of each sensor and data processing are done inside of UAS, using specific embedded computer and boards, defined as payload control board, to processing data onboard and reduce the telemetry throughput, as sample, the imagery control, image processing and pattern recognition, navigation avoid obstacle, and to control the Drone in automatic or remotely piloted mode, giving some 'intelligence' for the squadron operation (swarm), using the technic of area saturation to inspect all local infrastructure. The main characteristic of our application is the management data and the visual process. In this scope the aerial platform has embedded a set of cameras able to collect the image of the electrical asset, e.g., power line poles, lightning rod, rust/corrosion, damaged bolts and others, using thermal or regular cameras. The power line inspection process considers the Drone scanning the field, to identify the targets, doing the pattern recognition and other automatic procedures according to each operational scope and I.A. procedures, in order to increase the power lines maintenance process effectiveness. The Aerial systems are constantly monitored in real time by a set of Command and Control Stations (CCS), with dedicated infrastructure and software, from the field at the Situation room. This development is compounded by the management software (SIGI), UAS and specific IoT devices, totally integrated with the CCS room. Based on the data collected from the interest areas, SIGI software has the ability to show real time situational analysis, guiding the manager at resources optimization (material and human) improving the efficiency of resource allocation in these areas, reducing process costs and improving the maintenance management process improving the Brazilian energy services.