José Miguel de Almeida Pedro

I am an industrial Ph.D. student supervised by Patric Jensfelt, José Araújo and André Mateus. At Ericsson Research, I am in the Sensing & Perception team. At KTH, I am at the Division of Robotics, Perception and Learning (RPL) in the School of Electrical Engineering and Computer Science (EECS). I am funded by the Wallenberg AI, Autonomous Systems and Software Program (WASP).

My research interests span Robotics, Computer Vision and AI, with a special focus on Simultaneous Localisation and Mapping (SLAM). I'm most interested in understanding how map representations shape the properties of SLAM. Currently, I'm working on camera pose estimation in rendering-based frameworks, such as Neural Radiance Fields (NeRFs) or 3D Gaussian Splatting.

I am also currently co-supervising two Master's thesis projects:

- Diogo Paulo, Self-supervised Novel View Synthesis for Visual Localization & Mapping (co-supervised with Paula Carbó Cubero).

- Montserrat Cuturrufo, Robust Cycle-based Pose-Graph Optimization (co-supervised with David Umsonst).

Together with Ludvig Ericson, we explored some long-standing questions regarding autonomous exploration planning. The resulting work,"Information Gain is Not All You Need", was presented at ECMR 2025 in Padua.

If you're a Master's student and you're interested in doing a thesis project with me, reach out!