Niki Trigoni

About

Niki Trigoni is a prominent computer scientist whose research sits at the intersection of autonomous systems, deep learning, and spatial intelligence — with particular focus on localization, mapping, and navigation for robots and mobile devices. Her work has significantly advanced the field of visual SLAM, structure from motion, and odometry, most notably through pioneering deep learning-based approaches that move beyond traditional hand-crafted algorithms. Her 2018 survey on visual SLAM in dynamic environments has accumulated nearly 400 citations, while her influential work on sequence-to-sequence probabilistic visual odometry demonstrated how deep neural networks could reframe classical robotics problems with remarkable effectiveness. Trigoni has also made substantial contributions to pedestrian inertial navigation, thermal-inertial odometry, and mmWave radar-based sensing — including novel health monitoring applications such as robot-mounted heart rate detection. Her research spans both methodological innovation and real-world deployment, addressing challenging conditions like low visibility and GPS-denied environments. Through multiple high-impact surveys on deep learning for localization and mapping, she has helped define the trajectory of an emerging field, cementing her reputation as a leading voice in what she compellingly terms "spatial machine intelligence."

Research Focus

Key Achievements

Top Papers

1. 1
   Visual SLAM and Structure from Motion in Dynamic Environments

   391 citations · 2018
2. 2
   End-to-end, sequence-to-sequence probabilistic visual odometry through deep neural networks

   244 citations · 2017
3. 3
   Deep-Learning-Based Pedestrian Inertial Navigation: Methods, Data Set, and On-Device Inference

   154 citations · 2020
4. 4
   Towards Monocular Vision based Obstacle Avoidance through Deep Reinforcement Learning

   136 citations · 2017
5. 5
   milliEgo

   130 citations · 2020
6. 6
   A Survey on Deep Learning for Localization and Mapping: Towards the Age of Spatial Machine Intelligence

   106 citations · 2020
7. 7
   Learning with Training Wheels: Speeding up Training with a Simple Controller for Deep Reinforcement Learning

   83 citations · 2018
8. 8
   DeepTIO: A Deep Thermal-Inertial Odometry With Visual Hallucination

   82 citations · 2020
9. 9
   Heart Rate Sensing with a Robot Mounted mmWave Radar

   70 citations · 2020
10. 10
    Deep Learning for Visual Localization and Mapping: A Survey

    69 citations · 2023

Key Collaborators

Contact & Links