An OctoMap-based 3D CostMap

The development of OctoMap created an efficient, discrete, three-dimensional (3D) mapping framework that has been very useful in autonomous robotics for representing occupied and free space in the robot's environment. This paper presents an OctoMap-based CostMap, with cell classifications to represent obstacle safety zones, exploration frontiers, and WiFi status, by rebranding free space with this information. The CostMap inherits OctoMap features, including efficient memory usage, and parent cells retain the set of cell types for all child cells, allowing efficient searching. Applications include aerial vehicle path planning. CostMap formation and an example in 3D motion planning are given.