Drone swarms work together to improve sensing and communication
A new framework uses drone swarms to better match resources to vehicle demand across different regions.
A new framework allows drone swarms to cooperate across different regions. This helps with integrated sensing and communication (ISAC) in air-ground networks.
The system uses hierarchical multi-agent reinforcement learning (H-MARL) to coordinate decisions. This approach uses slow-timescale inter-region UAV reallocation and fast-timescale intra-region UAV control.
UAV reallocation reduces service disparities by matching resources to vehicle demand. Adaptive handshaking guided by marginal Cramér--Rao bound (CRB) utilities balances phase synchronization gains against handshaking energy consumption.
The framework combines observations from different regions to increase observation diversity. It also optimizes UAV positions and transmit power within each region to improve performance.
Reallocation can improve quality of service (QoS) by over 20 percentage points when resource-demand mismatches are severe. Cooperative sensing with adaptive handshaking incurs only about a 5.0% increase in mean CRB relative to perfect synchronization.
Key facts
- Reallocation can improve quality of service (QoS) by over 20 percentage points.
- Cooperative sensing with adaptive handshaking has a 5.0% increase in mean CRB relative to perfect synchronization.
- The framework uses hierarchical multi-agent reinforcement learning (H-MARL) for coordination.