<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Théo Alessandro Hermann</title><description>Researcher and engineer. Measuring environments that resist measurement: informal urban form in Rio, tree canopy from aerial LiDAR, and perception on real robots.</description><link>https://theoalessandro.com/</link><language>en-us</language><item><title>ROS 2 fleet log triage</title><link>https://github.com/thrmnn/ros2-localization-triage</link><guid isPermaLink="true">https://github.com/thrmnn/ros2-localization-triage</guid><description>Robots that navigate from their own sensors fail quietly: the estimate drifts while the reported confidence stays high. I built four detectors that find those moments in the logs a fleet already records, and graded them against labelled recordings from five platforms.</description><pubDate>Thu, 27 Aug 2026 00:00:00 GMT</pubDate></item><item><title>A harness for AI agents</title><link>https://theoalessandro.com/projects/agent-harness/</link><guid isPermaLink="true">https://theoalessandro.com/projects/agent-harness/</guid><description>Agentic systems fail the way robots do: a tool call breaks, nothing traces it, and nobody notices until a user does. I built a typed tool layer with one structured trace line per call and a regression suite wired into CI, so a bad prompt change fails the build instead of the user.</description><pubDate>Fri, 12 Jun 2026 00:00:00 GMT</pubDate></item><item><title>A real-time vision pipeline</title><link>https://github.com/thrmnn/perception-pipeline</link><guid isPermaLink="true">https://github.com/thrmnn/perception-pipeline</guid><description>Four years of one-off perception prototypes, consolidated into one modular stack: detection, multi-object tracking, re-identification and pose, each swappable, with a latency budget the whole thing is held to.</description><pubDate>Mon, 08 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Morphometrics of informal urban form</title><link>https://senseablerio.mit.edu/</link><guid isPermaLink="true">https://senseablerio.mit.edu/</guid><description>Favelas are built without plans, and the descriptors urban analysis relies on assume a street grid that is not there. I build the ones that hold on this fabric, from terrain models and building footprints, fine enough to resolve a single alley.</description><pubDate>Wed, 01 Jan 2025 00:00:00 GMT</pubDate></item><item><title>Leaf area from an aerial scan</title><link>https://www.ams-institute.org/</link><guid isPermaLink="true">https://www.ams-institute.org/</guid><description>A city knows where its trees are, not how much leaf they carry. I recover leaf area per tree by inverting how much sky the canopy blocks in an aerial LiDAR survey the city already flies for other reasons.</description><pubDate>Sun, 01 Sep 2024 00:00:00 GMT</pubDate></item><item><title>Camera perception for canal boats</title><link>https://theoalessandro.com/projects/autonomous-boat-perception/</link><guid isPermaLink="true">https://theoalessandro.com/projects/autonomous-boat-perception/</guid><description>Roboat&apos;s autonomy ran on LiDAR. As its first intern I added the camera side: detection and tracking on the vessel&apos;s own edge hardware, and the MLOps to keep models and data versioned as the recordings grew.</description><pubDate>Thu, 01 Aug 2024 00:00:00 GMT</pubDate></item><item><title>Event Cameras for High-Speed Drone State Estimation</title><link>https://www.iri.upc.edu/</link><guid isPermaLink="true">https://www.iri.upc.edu/</guid><description>6-DOF pose estimation through flight maneuvers exceeding 5g acceleration, where conventional frame cameras fail.</description><pubDate>Sat, 01 Jul 2023 00:00:00 GMT</pubDate></item><item><title>Autonomous Mobile Robot Navigation — Segway Loomo</title><link>https://www.epfl.ch/labs/vita/</link><guid isPermaLink="true">https://www.epfl.ch/labs/vita/</guid><description>A full autonomy stack — SLAM, planning, obstacle avoidance, person-following — validated end-to-end on a physical robot (Segway Loomo, EPFL VITA Lab research platform).</description><pubDate>Thu, 01 Jun 2023 00:00:00 GMT</pubDate></item></channel></rss>