Papers
2
Total Citations
41
H-Index
2
About
An Zou is a leading researcher at the intersection of real-time systems, heterogeneous computing, and cyber-physical systems, with a focus on enabling safe and efficient AI-driven autonomy. His work addresses the critical challenge of guaranteeing timing predictability for computationally intensive parallel workloads—such as deep neural networks—on modern heterogeneous platforms. Zou’s most cited paper, “RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks With Fine-Grain Utilization” (2023, 39 citations), introduces a novel scheduling framework that ensures hard real-time guarantees for GPU-accelerated tasks, a foundational contribution for safety-critical applications in autonomous vehicles and robotics. In his more recent work, “SCENIC: Capability and Scheduling Co-Design for Intelligent Controller on Heterogeneous Platforms” (2024), Zou pioneers a co-design approach that simultaneously optimizes hardware capability and scheduling policies for intelligent controllers, moving beyond traditional homogeneous-platform assumptions. This work is particularly impactful for modern control systems where DNN-based controllers must meet stringent real-time constraints. Through these contributions, Zou is shaping the future of predictable, high-performance computing in autonomous systems, bridging the gap between cutting-edge AI and the rigorous demands of real-time operation.
Research Focus
Key Achievements
Top Papers
- 1
- 2