Adaptive Heuristics for Scheduling DNN Inferencing on Edge and Cloud for Personalized UAV Fleets
Suman Raj, Radhika Mittal, Harshil Gupta, Yogesh Simmhan
TL;DR
This work tackles deadline-sensitive DNN inferencing for fleets of buddy drones aiding visually impaired people, by proposing DEMS-A, a deadline-driven edge+cloud scheduler, and GEMS, a QoE-guaranteeing extension. DEMS-A dynamically migrates tasks between edge and cloud, leverages work-stealing, and adapts to network variability to maximize task on-time completion and QoS utility, while GEMS ensures user-facing QoE by maintaining completion rates within tumbling windows. The approach is validated on realistic, hardware-backed VIP workloads with Jetson and AWS Lambda, showing up to 88% task completion, up to 2.7x QoS gains, and substantial QoE improvements, including field validation with real drones. The results demonstrate a scalable, end-to-end VIP navigation platform with practical implications for latency-sensitive drone analytics and can be extended to other latency-critical drone applications and mobile edge-cloud scenarios.
Abstract
Drone fleets with onboard cameras coupled with computer vision and DNN inferencing models can support diverse applications. One such novel domain is for one or more buddy drones to assist Visually Impaired People (VIPs) lead an active lifestyle. Video inferencing tasks from such drones can help both navigate the drone and provide situation awareness to the VIP, and hence have strict execution deadlines. We propose a deadline-driven heuristic, DEMS-A, to schedule diverse DNN tasks generated continuously to perform inferencing over video segments generated by multiple drones linked to an edge, with the option to execute on the cloud. We use strategies like task dropping, work stealing and migration, and dynamic adaptation to cloud variability, to guarantee a Quality of Service (QoS), i.e. maximize the utility and the number of tasks completed. We also introduce an additional Quality of Experience (QoE) metric useful to the assistive drone domain, which values the frequency of success for task types to ensure the responsiveness and reliability of the VIP application. We extend our DEMS solution to GEMS to solve this. We evaluate these strategies, using (i) an emulated setup of a fleet of over 80 drones supporting over 25 VIPs, with real DNN models executing on pre-recorded drone video streams, using Jetson Nano edges and AWS Lambda cloud functions, and (ii) a real-world setup of a Tello drone and a Jetson Orin Nano edge generating drone commands to follow a VIP in real-time. Our strategies present a task completion rate of up to 88%, up to 2.7x higher QoS utility compared to the baselines, a further 16% higher QoS utility while adapting to network variability, and up to 75% higher QoE utility. Our practical validation exhibits task completion of up to 87% for GEMS and 33% higher total utility of GEMS compared to edge-only.