Tianyu Jia

About

Tianyu Jia is a researcher at the forefront of ensuring safety and resilience in autonomous systems, with a focus on learning-based navigation, micro aerial vehicles (UAVs), and edge computing. His major contributions include developing end-to-end fault analysis frameworks like MAVFI and ROSFI, which assess the impact of silent data corruption (SDC) on mission-critical metrics such as flight time and success rate, and propose anomaly detection and recovery mechanisms. Jia also pioneered OMU, a probabilistic 3D occupancy mapping accelerator that enables real-time OctoMap at the edge, addressing compute and memory constraints for autonomous machines. His work, cited over 60 times, has been published in top venues, with his 2021 paper on fault tolerance in learning-based navigation systems garnering 22 citations. Notably, Jia’s research bridges hardware and software resilience, offering practical solutions for drones, robots, and vehicles. His achievements highlight a commitment to making autonomous systems safer and more reliable, making him a key voice in the field of dependable AI and robotics.

Research Focus

Key Achievements

Top Papers

1. 1
   Analyzing and Improving Fault Tolerance of Learning-Based Navigation Systems

   22 citations · 2021
2. 2
   MAVFI: An End-to-End Fault Analysis Framework with Anomaly Detection and Recovery for Micro Aerial Vehicles

   17 citations · 2023
3. 3
   OMU: A Probabilistic 3D Occupancy Mapping Accelerator for Real-time OctoMap at the Edge

   10 citations · 2022
4. 4
   Silent Data Corruption in Robot Operating System: A Case for End-to-End System-Level Fault Analysis Using Autonomous UAVs

   6 citations · 2023
5. 5
   MAVFI: An End-to-End Fault Analysis Framework with Anomaly Detection and Recovery for Micro Aerial Vehicles

   5 citations · 2021
6. 6
   OMU: A Probabilistic 3D Occupancy Mapping Accelerator for Real-time OctoMap at the Edge

   2 citations · 2022

Key Collaborators

Contact & Links