2025-05-05
Deep reinforcement learning six-pumping UAV
The technology of Unmanned Aerial Vehicles (UAVs), or simply drones, has gained much attention over the past two decades, with a rapid decrease in cost and increase in potential applications such as surveillance, journalism, agriculture, film-making, entertainment, etc. With the growth in the popularity of First-person view (FPV) drone piloting comes the challenge of transmitting control information and streaming video at very low latencies to enable proper real-time control. Due to strict latency requirements, FPV technology stagnated, with analog video transmission currently representing the most common implementation. In this paper, we present our preliminary research directed towards developing a digital alternative to this technology. We report on a user study designed to assess the Quality of Experience (QoE) of an FPV system coupled with a cloud-based drone flight simulator, with the focus being on video quality in ideal network conditions. Obtained results show that video encoding parameters have a significant impact on user perceived QoE. Furthermore, our findings highlight the relevance of simulator sickness in this kind of system.
The technology of Unmanned Aerial Vehicles (UAVs), or simply drones, has gained much attention over the past two decades, with a rapid decrease in cost and increase in potential applications such as surveillance, journalism, agriculture, film-making, entertainment, etc. With the growth in the popularity of First-person view (FPV) drone piloting comes the challenge of transmitting control information and streaming video at very low latencies to enable proper real-time control. Due to strict latency requirements, FPV technology stagnated, with analog video transmission currently representing the most common implementation. In this paper, we present our preliminary research directed towards developing a digital alternative to this technology. We report on a user study designed to assess the Quality of Experience (QoE) of an FPV system coupled with a cloud-based drone flight simulator, with the focus being on video quality in ideal network conditions. Obtained results show that video encoding parameters have a significant impact on user perceived QoE. Furthermore, our findings highlight the relevance of simulator sickness in this kind of system.
.png){kind=small}