Safe swarm navigation in cluttered indoor environment requires long-horizon planning, reactive obstacle avoidance, and adaptive compliance. We propose ImpedanceDiffusion, a hierarchical framework that leverages image-conditioned diffusion-based global path planning with Artificial Potential Field (APF) tracking and semantic-aware variable impedance control for aerial drone swarms. The diffusion model generates geometric global trajectories directly from RGB images without explicit map construction. These trajectories are tracked by an APF-based reactive layer, while a VLM-RAG module performs semantic obstacle classification with 90% retrieval accuracy to adapt impedance parameters for mixed obstacle environments during execution. Two diffusion planners are evaluated: (i) a top-view long-horizon planner using single-pass inference and (ii) a first-person-view (FPV) short-horizon planner deployed via a two-stage inference pipeline. Both planners achieve a 100% trajectory generation rate across twenty static and dynamic experimental configurations and are validated via zero-shot sim-to-real deployment on Crazyflie 2.1 drones through the hierarchical APF-impedance control stack. The top-view planner produces smoother trajectories that yield conservative tracking speeds of 1.0-1.2 m/s near hard obstacles and 0.6-1.0 m/s near soft obstacles. In contrast, the FPV planner generates trajectories with greater local clearance and typically higher speeds, reaching 1.4-2.0 m/s near hard obstacles and up to 1.6 m/s near soft obstacles. Across 20 experimental configurations (100 total runs), the framework achieved a 92% success rate while maintaining stable impedance-based formation control with bounded oscillations and no in-flight collisions, demonstrating reliable and adaptive swarm navigation in cluttered indoor environments.
Figures & tables
Fig. 1 : ImpedanceDiffusion in a mixed hard (rigid poles) and soft (human) obstacle environment. Diffusion-based planning generates the global path, APF ensures reactive avoidance (red dashed), and adaptive impedance links maintain formation cohesion and compliant interaction during flight.
Fig. 2 : System architecture of ImpedanceDiffusion. A top-view image and user prompt are processed by Molmo to identify obstacle type. The VLM–RAG module retrieves the corresponding impedance parameters from a custom database. In parallel, the image is provided to the diffusion model to generate a global trajectory, which is executed using APF-based tracking with adaptive inter-drone and drone–obstacle impedance control.
Drone–Drone Parameters
Cyl.
Chair
Trolley
Gate
Human
Virtual Mass (kg)
1
1
0.8
1
5
Virtual Stiffness (N/m)
7
7
7
7
1
Virtual Damping (Ns/m)
3
3
3
3
2
Drone–Obstacle Parameters
Cyl.
Chair
Trolley
Gate
Human
Virtual Mass (kg)
1
0.8
0.8
1.2
1
Virtual Stiffness (N/m)
9
10
5
8
16
TABLE I : Experimentally Validated Safe Virtual Impedance Link and Path Parameters for Drone–Drone and Drone–Obstacle Interactions, Stored in the Custom Database for Each Obstacle Type
Fig. 3 : Static environment with hard obstacles and a gate. Rigid compliance with small deflection is observed near hard objects.
Fig. 4 : Static cluttered environment with hard obstacles and one human. Soft compliance near the human results in larger deflection and reduced speed.
Fig. 5 : Dynamic environment with one moving human and one hard obstacle. Soft compliance near the human and rigid compliance near the chair are observed.
Fig. 6 : Dynamic scenario with two humans. Soft compliance with large deflection is observed.
Exp.
Scenario
Planner
Path
Coll.
Goal
Turn
(m)
(-)
(m)
(rad)
E1
Cyl.–Gate
P1
2.013
0.186
0.026
10.541
P2
2.191
0.067
0.028
7.279
E2
2 Cyl.–Gate
P1
2.148
0.352
0.044
9.254
P2
2.235
0.146
0.047
9.022
E3
Cyl.–Gate–Chair
P1
2.093
0.338
0.016
11.333
TABLE II : Performance Comparison Across Eight Experimental Scenarios
Aerial robotics studies the design, perception, control, and coordination of flying robots for navigation, inspection, transport, and physical interaction.
Vision-based UAV navigation becomes challenging when navigation targets are distributed across complementary camera views and cannot be reliably observed from a single viewpoint. We propose AgenticDiffusion, an agentic multi-view UAV navigation framework that semantically coordinates first-person-view (FPV) and top-view observations for mission-level navigation. Given a natural-language instruction, AgenticDiffusion identifies the requested targets, selects the most appropriate camera view for each navigation task, determines the corresponding navigation goal, and invokes the appropriate view-conditioned diffusion planner for trajectory generation. The resulting trajectories are executed using Nonlinear Model Predictive Control (NMPC). AgenticDiffusion was evaluated in four real-world indoor scenarios, achieving an overall mission success rate of 80% across 40 physical-flight trials. In mixed-visibility scenarios, where the requested targets were distributed across FPV and top-view observations, coordinated multi-view navigation reduced average mission time by 50.8% relative to FPV-only navigation and by 26.8% relative to Top-only navigation. The semantic view-selection mechanism was also robust to lexical variation in target descriptions, achieving 100% accuracy across 66 test cases, compared with 63.64% for a confidence-based view-selection baseline. In a substantially larger Gazebo environment, AgenticDiffusion achieved a 90% mission success rate and completed the multi-stage mission, whereas the FPV-only and Top-only variants were unable to complete all requested navigation tasks.
Faryal Batool, Muhammad Ahsan Mustafa, Fawad Mehboob +2
Autonomous multi-drone navigation requires fleets to service distributed objectives in cluttered environments under tight computational budgets. Efficient coordination depends on task bundling, where each drone visits multiple objectives along its route. Separate solvers for cost estimation, assignment, and execution incur redundant graph search and produce long, abrupt paths. We propose FlockDiffusion, a learned framework combining a scene graph encoder, an explicit allocation head, an assignment conditioned diffusion transformer, and a closed form trajectory decoder. An autoregressive teacher provides offline supervision for parallel fleet trajectory generation. PyBullet ablations show that bundling increases task completion from 50% to 100%, while our complete teacher further reduces route cost by 8.4% relative to MAGNNET with bundling. In the optimized scalability benchmark, evaluated on 100 scenes per density with ten drones, FlockDiffusion achieves 6.2 to 7.6 times faster inference and approximately 37% shorter routes than the classical pipeline. As nominal task counts increase from 20 to 40, latency rises from 7.8 to 11.1 ms, compared with 48.0 to 75.8 ms for the baseline. In a separate evaluation across five Gazebo environments, FlockDiffusion achieves 100% planner coverage and reduces planned route cost by 15.4% relative to the baseline with bundling. These results demonstrate efficient planning under increasing task density in configurations that are demanding to reproduce with physical drone fleets.
Iana Zhura, Satenik Akopyan, Roohan Ahmed Khan +3
Skolkovo Institute of Science and Technology, Moscow, Russia
Accurate 4D trajectory prediction and closed-loop tracking are essential for Unmanned Aerial Vehicle (UAV) swarms to achieve safe and efficient operations in complex low-altitude environments such as urban airspaces, industrial sites, and indoor facilities. However, this task remains challenging due to intrinsic nonlinearity of UAV swarm dynamics and strict real-time constraints of swarm formation control. To address these challenges, we propose a unified framework that couples coarse-to-fine trajectory forecasting with uncertainty-aware Distributed Nonlinear Model Predictive Control (DNMPC). Our approach features two key innovations: 1) a dimension-decoupled trajectory prediction module that reduces computational complexity by forecasting axis-wise motion, and 2) a diffusion-based residual dynamics refinement module that captures temporally correlated dynamic uncertainties. These refined predictions are then integrated into a DNMPC loop to ensure formation stability. We also introduce a synchronized multi-scenario 4D UAV swarm dataset spanning six representative airspace scenarios. The dataset contains over 7,900 frames of synchronized three-UAV trajectories with frame-level annotations of speed intention and target sector. Extensive experiments demonstrate that our approach outperforms state-of-the-art baselines, reducing trajectory tracking error by up to 10-15% and achieving sub-0.07,m average tracking error in complex urban and industrial environments, while maintaining real-time inference speeds of 34 FPS (sub-30 ms latency) suitable for agile flight.
Tianshun Li, Hongliang Lu, Haoang Li +1
The Hong Kong University of Science and Technology (Guangzhou), China · Southern University of Science and Technology, China and MoSense Technologies, China