Drone Gesture Control Research
Shows physical AI roots in human-machine control, sensing constraints, and lightweight robotics.
system visual
Protected or source media unavailable. Architecture preview used instead.
System Overview
Published Crazyflie drone control work using MediaPipe gesture and body-pose recognition under low-light constraints.
Publication/research.
Implementation Signals
System flow
A compact view of how inputs move through processing, orchestration, validation, and output.
Monochrome Camera
MediaPipe Landmarks
Gesture / Pose
Command Mapping
Drone Response
Engineering decisions
Decision Record
Use landmark-based control
Problem: Low-light drone interaction makes image classification brittle.
Approach: Use MediaPipe landmarks and pose/gesture recognition.
Tradeoff: Requires reliable landmark detection.
Outcome: LinkedIn/publication summary states landmark models were superior to traditional image classification.
Results / Learnings
Publication
Springer Nature Singapore, ICEEE 2024.
Finding
Single landmark-based model worked across brightness levels and colors per public summary.
Progressive depth
This page keeps the outcome and architecture visible first. Implementation stack, decisions, constraints, and media are available below so technical depth is opt-in rather than forced.
Adjacent systems
ROV2019 Navigation Stack
Underwater ROV navigation and hardware repository with PC-to-Arduino communication, camera configs, and engineering docs.
Repo: Includes navigation, arms, Arduino libraries, MQTT demo, camera configs, and engineering docs.
OpenVelocity Estimation with Optical Flow
Vehicle speed estimation system using classical optical flow, RAFT, YOLO, Raspberry Pi, and camera-stream constraints.
Repo: Package notes document RAFT, YOLO, KITTI evaluation, and Pi Camera streaming.
Open