Research Areas: Difference between revisions

From Innovation
Jump to: navigation, search
Created page with "== Research Areas == Our work on system software unfolds along three intertwined directions. === 🧠 AI & Intelligent Software === Bringing AI to the heart of system software. <div style="display:flex; flex-wrap:wrap; gap:15px; margin:15px 0;"> <div style="flex:1 1 280px; padding:20px; background:#faf5ff; border-left:4px solid #8a3aa8; border-radius:6px;"> <div style="font-size:115%; font-weight:bold; color:#4a1a6c; margin-bottom:8px;"> 🤖 AI-Driven System Softwa..."
 
No edit summary
Line 1: Line 1:
__NOTOC__
== Research Areas ==
== Research Areas ==



Revision as of 21:37, 3 May 2026

Research Areas

Our work on system software unfolds along three intertwined directions.

🧠 AI & Intelligent Software

Bringing AI to the heart of system software.

🤖 AI-Driven System Software

Embedding AI into the system software stack to enable self-adaptive, self-diagnostic, and intelligent decision-making systems. This includes applying Vision-Language Models (VLM) and Agentic AI at the system level.

🦾 Physical AI

Designing and verifying AI systems that interact directly with the physical world through sensors and actuators — ensuring that intelligence operates safely and predictably in autonomous platforms.

✍️ LLM for Software Engineering

Leveraging Large Language Models to transform software engineering workflows — from requirements analysis and code generation to automated documentation of safety-critical software.

🛡️ Reliability & Assurance

Verification and monitoring techniques that guarantee safety and reliability.

🛡️ Runtime Assurance (RTA)

Designing runtime monitoring and fail-safe mechanisms that guarantee safety of aerospace and autonomous systems during operation, with verification methodologies aligned with avionics standards such as DO-178C.

🔗 Multi-Agent Systems (MAS)

Algorithms for coordination and cooperation among multiple autonomous agents, along with verification methods for distributed systems — ensuring both safety and mission success when multiple unmanned assets operate together.

🏗️ Structural Health Monitoring (SHM)

Sensor-driven techniques for diagnosing and predicting structural conditions, with applications in aircraft structural health monitoring — including AI-based anomaly detection and predictive maintenance.

⚙️ Systems & Infrastructure

The foundational layer — system software infrastructure and distributed computing.

🛰️ Digital Twin (DT)

Building digital twin models with real-time synchronization for aircraft, UAVs, and autonomous systems — enabling monitoring, predictive diagnostics, and simulation-driven decision-making.

☁️ Edge Computing & Computation Offloading

Optimizing the distribution of computation between resource-constrained edge devices and cloud/server backends — balancing latency, power, and communication costs in distributed computing architectures.

✈️ UAV & Airborne Software

Designing and verifying embedded software for unmanned aerial vehicles — from flight control to mission management — including work on open platforms such as PX4 and ROS2 and applications of avionics software standards.

→ Explore our research in depth