Research map / Reinforcement learning / Visual and representation RL
UAV$^2$: A Unified and Adaptive Scheduling Framework for UAV Autopilot System with Reinforcement Learning
Zeying Li, shuai zhao, Chaowen Wu, Boyang Li, Kai Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add deviation penalties to the reward (ablation) | UAV2 reward · HIL Pillars | Success rate (%): 42 -> 86 | Improved |
| add deviation penalties to the reward (ablation) | UAV2 reward · HIL Pillars | Mean deviation (m): 0.35 -> 0.15 | Improved |
| add CPU penalty to the reward (ablation) | UAV2 reward · HIL Pillars | Mean CPU time (ms): 83.15 -> 73.99 | Improved |
| add CPU penalty to the reward (ablation) | UAV2 reward · HIL Pillars | Success rate (%): 66 -> 86 | Improved |
| add collision penalty to the reward (ablation) | UAV2 reward · HIL Pillars | Success rate (%): 28 -> 86 | Improved |
| scale all reward weights jointly by 0.1 or 10 (ablation) | UAV2 reward · HIL Pillars | Success rate (%): 84% and 86% vs 86% | No clear effect |
| reassign weight magnitudes across reward terms (ablation) | UAV2 reward · HIL Pillars | Success rate (%): 86 -> 54, 50, 30, 32, 26 | Hurt |
| use learned scheduling policy instead of fixed-rate scheduling (vs strategy A) | UAV2 · Real-world flights | Success rate (%): 40% -> 100% | Improved |
| use learned scheduling policy instead of fixed-rate scheduling | UAV2 · HIL Pillars | Success rate (%): highest success rate (86.0%) | Improved |
Closest papers by meaning
- UAV-Flow Colosseo: A Real-World Benchmark for Flying-on-a-Word UAV Imitation Learning (NeurIPS 2025 · similarity 0.84)
- Improving Autonomous Separation Assurance through Distributed Reinforcement Learning with Attention Networks (AAAI 2024 · similarity 0.83)
- NeuralPlane: An Efficiently Parallelizable Platform for Fixed-wing Aircraft Control with Reinforcement Learning (NeurIPS 2024 · similarity 0.83)
- U2UData+: A Scalable Swarm UAVs Autonomous Flight Dataset for Embodied Long-horizon Tasks (AAAI 2026 · similarity 0.83)
- DesignX: Human-Competitive Algorithm Designer for Black-Box Optimization (NeurIPS 2025 · similarity 0.82)
- TGPO: Temporal Grounded Policy Optimization for Signal Temporal Logic Tasks (NeurIPS 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
