R&D Project Track Record

---

A 6-DoF dynamics accelerator was implemented in the Programmable Logic (PL) section of a System-on-Chip (SoC), enabling the parallel execution of 434 aerial vehicle models at a 1 kHz update rate. The resulting high-throughput IP core provided the computational foundation for swarm UAV emulation and Monte Carlo simulation.

Our team provided avionics engineering support for the development of a flight vehicle system, contributing to fin actuation, long-range data links, real-time computing hardware, power control electronics, and Real-Time Operating System (RTOS) applications. The work covered the integration of motor, motor driver, and sensor components within the fin actuation subsystem, together with supporting embedded hardware and software elements.

A morphing rotary-wing UAV was developed to stabilize itself and transition to controlled flight after pneumatic launch. The system was tested from both a stationary launcher and a moving vehicle traveling at 60 km/h. The project demonstrated compact deployment, rapid stabilization, and reliable vehicle-based launch capability for field applications.

A central route planner was developed to coordinate multiple UAVs through an extended star communication topology. The system supported in-flight target updates, airborne and ground-based launches, smooth rendezvous trajectories, and operator-controlled terminal approach timing. The project also included custom radio firmware development with an auto-relay handover capability, strengthening our experience in multi-UAV coordination and formation flight.

The FACT project explored next-generation Communication, Navigation, and Surveillance (CNS) technologies for future Air Traffic Management (ATM) and U-space integration. The work focused on performance-based CNS functions, cross-domain technology adaptation, and safe shared-airspace operations involving conventional aircraft and UAVs. In the final demonstration, a helicopter, an airplane, and two quadcopters operated safely in the same airspace while executing CNS functions.

Dynamic models of the quadcopters and fixed-wing UAVs in the fleet were obtained through system identification, enabling improvements in flight controller performance. The quadcopters achieved cruising speeds up to 30 m/s and thrust-to-weight ratios of up to 4:1. The project strengthened our experience in high-agility UAV modeling, GNSS-based precision targeting, GNSS uncertainty analysis, and RTK-assisted navigation.

Within a 4 km² operational area, the system enabled operator-designated surveillance of a region, point, or moving convoy. A central planning algorithm assigned missions to airborne fixed-wing UAVs and generated flight routes accordingly. At least three of six UAVs remained airborne at all times, providing continuous 2-axis gimbaled camera coverage and real-time video transmission to the ground station. When a UAV reached a critical energy level, it autonomously returned to base, while a replacement UAV took over the mission without interruption.

This first field-oriented project established our group’s operational UAV capability. Gasoline-powered fixed-wing UAVs with over one hour of endurance were deployed to locate a victim who could only transmit by pressing the push-to-talk button of a VHF radio operating at an unknown frequency. Equipped with Software Defined Radio (SDR) units and omnidirectional antennas, the UAVs measured RF signal strength, performed decentralized route planning, and collaboratively estimated the transmitter location. The system localized the radio within a 40 m deviation in a 25 km² test area.