Perception-aware Exploration for Consumer-grade UAVs

arXiv cs.RO / 4/14/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper extends autonomous multi-UAV exploration methods to consumer-grade drones (e.g., DJI Mini 3 Pro) by adding a perception-aware pipeline for selecting informative viewpoint pairs to estimate depth.
It couples depth estimation with trajectory planning that explicitly satisfies motion constraints needed for reliable odometry under consumer hardware limitations.
For multi-drone operation, the authors introduce a semi-distributed communication scheme designed to balance workload across UAVs while coordinating exploration.
The approach is evaluated in simulation across varying fleet sizes, demonstrating safe exploration and successful map reconstruction despite consumer-grade constraints, and includes proofs of safety and reconstructability.
Overall, the work advances practical autonomy for small UAVs by addressing both perception (depth) and systems concerns (motion/communication coordination) together.

Abstract

In our work, we extend the current state-of-the-art approach for autonomous multi-UAV exploration to consumer-level UAVs, such as the DJI Mini 3 Pro. We propose a pipeline that selects viewpoint pairs from which the depth can be estimated and plans the trajectory that satisfies motion constraints necessary for odometry estimation. For the multi-UAV exploration, we propose a semi-distributed communication scheme that distributes the workload in a balanced manner. We evaluate our model performance in simulation for different numbers of UAVs and prove its ability to safely explore the environment and reconstruct the map even with the hardware limitations of consumer-grade UAVs.