Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Edge-Native, Behavior-Adaptive Drone System for Wildlife Monitoring

Jenna Kline, Rugved Katole, Tanya Berger-Wolf, Christopher Stewart

TL;DR

This work addresses the challenge of monitoring wildlife with drones without inducing stress that biases data. It introduces an edge-native, behavior-adaptive drone system that continuously tracks individual behaviors to compute a real-time group vigilance score and surfaces graduated alerts to the operator, enabling timely, context-aware interventions. The study demonstrates that edge-based inference (23.8 ms) meets the 33 ms frame-latency requirement, and retrospective analysis across seven missions shows substantial improvements in usable data and reductions in adverse behavior when vigilance-informed decisions are used versus manual piloting. The proposed human-on-the-loop decision-support approach—combining real-time vigilance monitoring with context-tunable thresholds—offers a practical pathway toward field-ready, scalable wildlife monitoring while preserving animal welfare. Future work aims to close the control loop with adaptive flight behavior, validated through live-field trials.

Abstract

Wildlife monitoring with drones must balance competing demands: approaching close enough to capture behaviorally-relevant video while avoiding stress responses that compromise animal welfare and data validity. Human operators face a fundamental attentional bottleneck: they cannot simultaneously control drone operations and monitor vigilance states across entire animal groups. By the time elevated vigilance becomes obvious, an adverse flee response by the animals may be unavoidable. To solve this challenge, we present an edge-native, behavior-adaptive drone system for wildlife monitoring. This configurable decision-support system augments operator expertise with automated group-level vigilance monitoring. Our system continuously tracks individual behaviors using YOLOv11m detection and YOLO-Behavior classification, aggregates vigilance states into a real-time group stress metric, and provides graduated alerts (alert vigilance to flee response) with operator-tunable thresholds for context-specific calibration. We derive service-level objectives (SLOs) from video frame rates and behavioral dynamics: to monitor 30fps video streams in real-time, our system must complete detection and classification within 33ms per frame. Our edge-native pipeline achieves 23.8ms total inference on GPU-accelerated hardware, meeting this constraint with a substantial margin. Retrospective analysis of seven wildlife monitoring missions demonstrates detection capability and quantifies the cost of reactive control: manual piloting results in 14 seconds average adverse behavior duration with 71.9% usable frames. Our analysis reveals operators could have received actionable alerts 51s before animals fled in 57% of missions. Simulating 5-second operator intervention yields a projected performance of 82.8% usable frames with 1-second adverse behavior duration,a 93% reduction compared to manual piloting.

Edge-Native, Behavior-Adaptive Drone System for Wildlife Monitoring

TL;DR

This work addresses the challenge of monitoring wildlife with drones without inducing stress that biases data. It introduces an edge-native, behavior-adaptive drone system that continuously tracks individual behaviors to compute a real-time group vigilance score and surfaces graduated alerts to the operator, enabling timely, context-aware interventions. The study demonstrates that edge-based inference (23.8 ms) meets the 33 ms frame-latency requirement, and retrospective analysis across seven missions shows substantial improvements in usable data and reductions in adverse behavior when vigilance-informed decisions are used versus manual piloting. The proposed human-on-the-loop decision-support approach—combining real-time vigilance monitoring with context-tunable thresholds—offers a practical pathway toward field-ready, scalable wildlife monitoring while preserving animal welfare. Future work aims to close the control loop with adaptive flight behavior, validated through live-field trials.

Abstract

Wildlife monitoring with drones must balance competing demands: approaching close enough to capture behaviorally-relevant video while avoiding stress responses that compromise animal welfare and data validity. Human operators face a fundamental attentional bottleneck: they cannot simultaneously control drone operations and monitor vigilance states across entire animal groups. By the time elevated vigilance becomes obvious, an adverse flee response by the animals may be unavoidable. To solve this challenge, we present an edge-native, behavior-adaptive drone system for wildlife monitoring. This configurable decision-support system augments operator expertise with automated group-level vigilance monitoring. Our system continuously tracks individual behaviors using YOLOv11m detection and YOLO-Behavior classification, aggregates vigilance states into a real-time group stress metric, and provides graduated alerts (alert vigilance to flee response) with operator-tunable thresholds for context-specific calibration. We derive service-level objectives (SLOs) from video frame rates and behavioral dynamics: to monitor 30fps video streams in real-time, our system must complete detection and classification within 33ms per frame. Our edge-native pipeline achieves 23.8ms total inference on GPU-accelerated hardware, meeting this constraint with a substantial margin. Retrospective analysis of seven wildlife monitoring missions demonstrates detection capability and quantifies the cost of reactive control: manual piloting results in 14 seconds average adverse behavior duration with 71.9% usable frames. Our analysis reveals operators could have received actionable alerts 51s before animals fled in 57% of missions. Simulating 5-second operator intervention yields a projected performance of 82.8% usable frames with 1-second adverse behavior duration,a 93% reduction compared to manual piloting.

Paper Structure

This paper contains 26 sections, 1 equation, 4 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background
  3. Automatic in-situ behavior recognition
  4. Human-autonomy teaming
  5. Vigilance monitoring for decision support
  6. System Design
  7. Service-Level Objectives for Behavior-Adaptive Navigation
  8. Vigilance Monitoring Algorithm
  9. Group Vigilance Score
  10. Context-Tunable Thresholds
  11. Uncertainty Handling
  12. Human-on-the-Loop Interface
  13. Evaluation Methodology
  14. Experimental Design
  15. Analysis Protocol
  16. ...and 11 more sections

Figures (4)

  • Figure 1: Edge-Native Behavior-Adaptive Drone Control for Wildlife Monitoring. The semi-autonomous drone system augments the operator's ability to observe the wildlife and react in time to prevent adverse behavior events. The ground-control station (GCS) provides the operator with information, allowing them to adjust the automatic, behavior-adaptive navigation appropriately. The GCS recieves video stream from the drone, and in turn sends commands for where the drone should fly next, using a group-tracking navigation policy kline2025wildwing, the wildlife monitor the human operator and the drone, and may flee if threatened.
  • Figure 2: Edge-Native, Behavior-Adaptive Drone System Diagram. (1) Real mission replay, (2) Edge-native navigation algorithm with group-tracking and behavior-adaptive flight, (3) Human-on-the-Loop user interface.
  • Figure 3: Experimental setup for replay - mission and sampling duration for representative video, with adverse behavior event.
  • Figure 4: Human-on-the-Loop User Interface for Behavior-Adaptive Drone System Monitoring and Vigilance Monitoring