Method of flight planning for airborne LiDAR using genetic algorithms

Abstract Conventional methods of flight planning for airborne LiDAR are heuristic in nature and use an iterative trial and error approach. A new system-based approach of flight planning is presented in this paper. The presented approach automatically derives flight planning parameters by minimizing the cost of data acquisition, which is represented by flight duration. The flight duration, which is the sum of the strip time and turning time, is minimized using genetic algorithms under the constraints of mapping requirements, hardware limitations, user-defined preferences, and various other requirements. The proposed approach is first validated for conventionally known test cases of regular shapes (rectangular and triangular). Thereafter, it is implemented for an arbitrarily shaped simulated test site with two commercially available airborne LiDAR sensors. Statistical results are presented for the above. Further, flight planning is performed for two real test sites. The demonstrated approach not only produces optimal results, but also avoids the assumptions of conventional methods. Furthermore, the approach requires the least amount of human intervention and, thus, eliminates the subjectivity that is imposed by individual flight planners for determining the flight planning parameters. Encouraged by these results, the authors suggest that the proposed approach can be further developed to include all possible components of flight planning in a future work.