Data Collection, Heat Map Generation for Crack Detection Using Robotic Dog Fused with FLIR Sensor

摘要

The imperative need to monitor critical infrastructure, encompassing power plant transformers and essential civil structures, is paramount for ensuring their sustained longevity and safety. A notable progression in this domain arises from the widespread availability of Forward-Looking Infrared (FLIR) cameras, made possible by reductions in manufacturing costs. These FLIR cameras, distinguished by their remarkable precision in temperature data collection and visualization, hold significant potential across diverse fields. This paper underscores the pioneering application of FLIR cameras in infrastructure monitoring, with a specific focus on power transformers and aging civil structures like bridges. In the realm of transformers, the FLIR camera exhibits the capability to delve into and record internal temperatures, facilitating the early identification of deviations from standard operational temperature ranges. Detection of such anomalies serves as a trigger for preemptive measures, thereby minimizing operational disruptions. Similarly, for structures such as bridges, routine FLIR analyses prove instrumental in discerning temperature variations that may indicate potential wear or fractures. This timely identification allows for proactive maintenance, ensuring structural integrity and averting potential hazards. To augment the safety and efficiency of data collection, this research advocates for the integration of FLIR cameras with mobile robots, exemplified by the Unitree GoAirl. This strategic amalgamation enables precise yet remote surveillance, mitigating risks to operators while maximizing the accuracy of infrastructure health assessments. The proposed approach not only advances the capabilities of FLIR technology but also her-alds a paradigm shift in the realm of infrastructure monitoring, promising enhanced reliability and safety standards.