Malcolm Mielle

Recent Publications

See 9 more →

CV

4 min read

Malcolm Mielle

Curriculum Vitae — last updated 2026. A PDF version is also available.

Chavannes-près-Renens, Switzerland | French/Canadian | LinkedIn | GitHub | Google Scholar | ORCID

Research profile

I’m a research team lead (R3) at the Schindler EPFL Lab, working at the intersection of machine learning, multimodal sensing, and robotics for the built environment. My research turns multi-sensor data into practical tools: thermal digital twins that guide building retrofit decisions, and AI-based safety systems for workers. I secured two Innosuisse grants as Principal Investigator, co-direct two PhD theses, and have a track record of taking research from idea to patent and deployment.

Work experience

Schindler, Lausanne, CH — Team Lead, Research (Jan 2022 – present)

  • Leading research on computer science, multi-modal sensor fusion, and robotics with a focus on built environment sustainability and worker safety.
  • Secured 2 Innosuisse grants as Principal Investigator (587k CHF total).
  • Supervised 10 Master theses and 16 internships — outcomes include 6 patents, 3 publications, 1 dataset, and 1 prize.
  • Project management (Agile methodology), hiring, and budget management.

Bluebotics, St-Sulpice, CH — R&D Software Engineer (July 2020 – Nov 2021)

  • Designed and implemented novel algorithms for real-time motion detection in 2D laser scans.
  • Implemented features that improved the installation time from multiple days to a couple of hours.
  • Agile project management, code review, and software architecture.

Education

PhD in Computer Science, Örebro University, Sweden (2015–2019)

Helping robots help us: Using prior information for localization, navigation, and human-robot interaction. Supervisors: Prof. Martin Magnusson and Prof. Achim J. Lilienthal.

Master of Engineering — Robotics, Polytech Paris Sorbonne, France (2011–2014)

Grants

As Principal Investigator (details on the Grants page, projects on the Projects page):

  • Safety Coach (Innosuisse, Jan 2025 – Jan 2027) — 269’334 CHF. AI-based companion leveraging wearable and wireless sensors to improve safety on construction sites. PI: Dr Malcolm Mielle, Dr Anisoara Ionescu (EPFL), Prof Andres Perez-Uribe (HEIG-VD). Partners: Schindler, EPFL, HEIG-VD.
  • INSULATED (Innosuisse, Jan 2024 – Jan 2027) — 317’814 CHF. Integrated solution for lean and abridged thermal evaluation with digital twins. PI: Dr Malcolm Mielle and Prof Olga Fink (EPFL). Partners: Schindler, EPFL.

Selected publications

Full list with DOIs, code, datasets, and BibTeX on the All publications page.

Journal articles:

  • Hassan, Forest, Fink, Mielle — ThermoNeRF: A multimodal Neural Radiance Field for joint RGB-thermal novel view synthesis of building facades, Advanced Engineering Informatics, 2025.
  • Xu, Mielle, Laborde, Waseem, Forest, Fink — Exploiting Semantic Scene Reconstruction for Estimating Building Envelope Characteristics, Building and Environment, 2025.
  • Mielle, Magnusson, Lilienthal — The Auto-Complete Graph: Merging and Mutual Correction of Sensor and Prior Maps for SLAM, Robotics, 2019.
  • Mielle, Magnusson, Lilienthal — URSIM: Unique Regions for Sketch Map Interpretation and Matching, Robotics, 2019.

Conference papers:

  • Skorokhodov, Xu, Fink, Mielle — SEAR: Bridging RGB and Thermal for Robust 3D Reconstruction, ECCV 2026.
  • Daher, Wang, Mielle — D-CAT: Decoupled Cross-Attention Transfer between Sensor Modalities for Unimodal Inference, ICRA 2026.
  • Chassaing, Forest, Fink, Mielle — Thermoxels: A Voxel-Based Method to Generate Simulation-Ready 3D Thermal Models, CISBAT 2025.
  • Sun, Mielle, Lilienthal, Magnusson — High-Fidelity SLAM Using Gaussian Splatting…, IROS 2024.
  • Sun, Mielle, Lilienthal, Magnusson — 3QFP: Efficient Neural Implicit Surface Reconstruction…, ICRA 2024.
  • Panchetti, Zheng, Bouri, Mielle — TEAM: A Parameter-Free Algorithm to Teach Collaborative Robots Motions from User Demonstrations, ICINCO 2023.
  • Mielle, Magnusson, Lilienthal — A Comparative Analysis of Radar and Lidar Sensing for Localization and Mapping, ECMR 2019.
  • Mielle, Magnusson, Lilienthal — MAORIS: Map of Ripples Segmentation, ICRA 2018.
  • Mielle, Magnusson, Andreasson, Lilienthal — SLAM Auto-Complete: Completing a Robot Map Using an Emergency Map, SSRR 2017.
  • Mielle, Magnusson, Lilienthal — Using Sketch-Maps for Robot Navigation: Interpretation and Matching, SSRR 2016.

Patents — 4 patent applications on elevator maintenance robotics and shaft measurement, see Patents.

Supervision and teaching

  • PhD co-direction: Chenghao Xu (EPFL, with Prof Olga Fink) — neural rendering and building characteristics estimation for building retrofit; Shuo Sun (Örebro University, with Dr Martin Magnusson and Prof Achim J. Lilienthal) — efficient mapping for mobile robots.
  • Master theses and internships: 10 Master theses, 16 internships — see Supervision.
  • Guest lectures: EPFL (machine learning for construction safety; computer vision for 3D reconstruction), ETH Zurich (machine learning for industrial application).
  • Teaching assistant: Örebro University — mathematics and introduction to Python (2016–2018).

Awards

  • Best Student Paper Award, IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), 2017 — for the Auto-Complete Graph.
  • ASCE Swiss Prize (2024), awarded to my Master student Mariam Hassan for her thesis on thermal reconstruction (ThermoNeRF).

Languages

French (mother tongue), English (C2), Swedish (A2).

Graph View

Backlinks