GoKawiilPurdue lab demonstrates GPS-free drone tracking a person through forest
Purdue computer science professor Aniket Bera's IDEAS Lab posted a video showing a quadcopter autonomously following a person through dense woods using only onboard computing, without GPS, maps, or cloud processing. The drone uses an occlusion-prediction method to estimate the target's path when trees block its view, and relies on open-source navigation tools including SUPER and FAST-LIO, originally developed by the University of Hong Kong's MaRS Lab, despite Bera's claim his team built the system from scratch.
GoKawiil's interpretation of the reporting above, not reported fact.
GPS signals often fail under thick forest canopy, so onboard autonomous navigation could be essential for drones operating in dense terrain where cloud connectivity and satellite positioning are unreliable. Purdue's lab has a $1.5 million, five-year contract with the Army Research Laboratory for GPS-independent navigation research, suggesting this demonstration may tie into military interest in autonomous terrain navigation, though the post doesn't confirm that funding link. The discrepancy over whether the software was built in-house versus adapted from existing open-source projects raises questions about how the achievement should be credited.
- Purdue's IDEAS Lab showed a drone tracking a person through forest with no GPS, maps, or cloud computing.
- The system uses occlusion prediction to estimate a target's movement when hidden by trees.
- Despite claims of building everything from scratch, the drone's displayed software uses open-source tools SUPER and FAST-LIO from Hong Kong's MaRS Lab.
DJI Neo Autonomous Mini Drone — Like the Purdue research drone, the DJI Neo can track and follow a subject autonomously without needing a controller, making it a fun consumer taste of the same follow-me tech. It's a great entry point for anyone intrigued by autonomous drone tracking who wants to experiment with similar vision-based following features.
See DJI Neo Autonomous Mini Drone on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: techspot.com — Zo Ahmed, 2026-10-28
Published there as: “Watch this autonomous drone chase a person through a forest without GPS, maps, or cloud computing”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.