With the continuous development of the power grid industry, the difficulty of substation operation and maintenance is also increasing. Because of its high mobility and low destructiveness, a rotary-wing unmanned aerial vehicle (UAV) can achieve effective detection and avoid the difficulty and poor accuracy of high-altitude inspection of substations. Owing to the ability of a UAV to integrate various sensors, such as visible light imaging sensors, thermal imaging sensors, and self-positioning sensors, it can ensure its own safety and that of the equipment to be tested during the inspection. Thus, in this paper, we study the application of rotary-wing multi-sensor UAVs in the high-altitude inspection of substations. Firstly, we establish a UAV inspection and autonomous detection model involving electromagnetic and other environmental influences. This model considers substation environmental restrictions, UAV physical constraints, and the limitations of sensor detection, and it is combined with a mobile edge computing platform. Secondly, we design an improved path planning algorithm combining the floating algorithm and the rapidly exploring random trees (RRT) algorithm and introduce improved strategies such as the Cauchy mutation strategy and Lèvy flight strategy to reduce the blindness of the algorithm search. Finally, by conducting simulation experiments and comparing the designed algorithm with traditional algorithms such as the ant colony algorithm, genetic algorithm, and simulated annealing algorithm, we verify the feasibility of the model and algorithm. The results of this study provide practical assistance to promote the automation and intelligence of substation inspection.