To solve the problems of existing power inspection robots deviating from their trajectory during inspection, incomplete coverage leading to missed inspections, and high redundancy leading to low efficiency, an inspection strategy based on multi-information fusion algorithms is proposed. The Cartographer algorithm for map construction is improved, the A* algorithm for global planning from the start point to the target endpoint is introduced, and the Teb algorithm for local path planning to avoid obstacles improved. By integrating feedback strategies from different mapping algorithms and parsing rules from different navigation models, a heuristic multi-information-adaptive rule is established. Experimental results show that the multi-information fusion algorithm performs path planning within a $135\mathrm{m}\times 115.5\mathrm{m}$ coordinate-unknown area in a typical State Grid-designed substation, with start point and target endpoint displacement deviation controlled within ±9.68%, heading angle deviation maintained within ±14.12%, inspection coverage ≥96%, and inspection redundancy ≤10%, meeting the needs of online robot inspection within a certain area.