S1S2 (System 1, System 2) blends classical robotics — the fast System 1 that drives the robot — with foundation models — the slow, deliberate System 2 that plans.https://s1s2.ai/https://s1s2.ai/2026/07/17/building-a-factor-graph-engine-from-scratch/https://s1s2.ai/2026/07/17/building-a-factor-graph-engine-from-scratch/Continuing my journey through the implementation of factor graph-based state estimation, we first explored how a classic Extended Kalman Filter (EKF) update is not a black-box matrix equation, but fundamentally a two-factor graph solve via Non-linear Least Squares (NLS). When we look under the hood Fri, 17 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/07/17/from-offline-to-online-slam/https://s1s2.ai/2026/07/17/from-offline-to-online-slam/Batch Factor Graphs unlocked fabulous offline SLAM to produce highly accurate maps. You collect data on a long robot run, feed it to the...Fri, 17 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/07/17/gaussians-and-the-information-form-the-secret-sauce-of-factor-graphs/https://s1s2.ai/2026/07/17/gaussians-and-the-information-form-the-secret-sauce-of-factor-graphs/When you first dive into State Estimation and Kalman Filters (whether Extended or Unscented), you learn that they belong to a broader family of Gaussian Filters. But soon enough, a close cousin is introduced: the Information Filter. This alternative view makes you realize that a Gaussian distributioFri, 17 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/07/13/lie-groups-and-on-manifold-optimization/https://s1s2.ai/2026/07/13/lie-groups-and-on-manifold-optimization/If you are a roboticist, you have certainly encountered issues when working with plain angles (e.g., θ in a 2D pose) and related rotations...Mon, 13 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/07/08/nonlinear-least-squares-one-gauss-newton-step-is-a-filter-many-are-a-smoother/https://s1s2.ai/2026/07/08/nonlinear-least-squares-one-gauss-newton-step-is-a-filter-many-are-a-smoother/In a previous breakdown, we explored how the classic Extended Kalman Filter (EKF) can be viewed as an optimized, local version of a factor graph, where an EKF update is essentially a single, two-factor graph solve. As we push deeper into Factor Graphs for State Estimation, SLAM, and auto-calibrationWed, 08 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/07/05/a-kalman-update-is-a-two-factor-factor-graph-solve/https://s1s2.ai/2026/07/05/a-kalman-update-is-a-two-factor-factor-graph-solve/The classical Kalman Filter was an engineering shortcut designed for an era when compute was severely limited and maintaining a history of past sensor measurements was too costly. It leveraged Bayes’ rule and the Markov property to exploit the fact that the present state only depends on the immediatSun, 05 Jul 2026 00:00:00 GMThttps://s1s2.ai/2026/06/30/making-your-state-estimator-ros-2-compliant-rep-105/https://s1s2.ai/2026/06/30/making-your-state-estimator-ros-2-compliant-rep-105/I have been having so much fun building and testing my custom state estimator. Although I am designing it to be framework-agnostic, I am building a dedicated ROS 2 layer. ROS 2 is the obvious choice for prototyping the System 1 / System 2 (Physical AI, Embodied AI, or whatever trendy term you preferTue, 30 Jun 2026 00:00:00 GMThttps://s1s2.ai/2026/06/25/fusing-lidar-odometry-for-gps-denied-navigation/https://s1s2.ai/2026/06/25/fusing-lidar-odometry-for-gps-denied-navigation/In a previous article, I described how GPS can be fused with pure dead reckoning to achieve more accurate outdoor robot pose estimation using an Extended Kalman Filter (EKF). But what if you are navigating indoors, or outdoors with a poor GPS signal? What if you are in a war zone where GPS is jammedThu, 25 Jun 2026 00:00:00 GMThttps://s1s2.ai/2026/06/24/augmenting-dead-reckoning-state-estimation-with-gps-for-better-odometry/https://s1s2.ai/2026/06/24/augmenting-dead-reckoning-state-estimation-with-gps-for-better-odometry/Relative sensor fusion can only take a mobile robot so far. To build truly robust autonomous systems, you have to bound dead reckoning drift with absolute global references. As part of my ongoing development of System 1 and System 2 robot autonomy infrastructure, I’ve expanded my Python Extended KalWed, 24 Jun 2026 00:00:00 GMThttps://s1s2.ai/2026/05/21/3d-perception-a-lidar-camera-pipeline-part-6-projecting-3d-intelligence-into-a-2d-map-as-a-birds-eye-view/https://s1s2.ai/2026/05/21/3d-perception-a-lidar-camera-pipeline-part-6-projecting-3d-intelligence-into-a-2d-map-as-a-birds-eye-view/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Thu, 21 May 2026 00:00:00 GMThttps://s1s2.ai/2026/05/20/3d-perception-a-lidar-camera-pipeline-part-5-merging-2d-intelligence-and-3d-depth-with-frustum-fusion/https://s1s2.ai/2026/05/20/3d-perception-a-lidar-camera-pipeline-part-5-merging-2d-intelligence-and-3d-depth-with-frustum-fusion/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Wed, 20 May 2026 00:00:00 GMThttps://s1s2.ai/2026/04/22/3d-perception-a-lidar-camera-pipeline-part-4-2d-camera-detections-with-yolo/https://s1s2.ai/2026/04/22/3d-perception-a-lidar-camera-pipeline-part-4-2d-camera-detections-with-yolo/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Wed, 22 Apr 2026 00:00:00 GMThttps://s1s2.ai/2026/04/08/3d-perception-a-lidar-camera-pipeline-part-3-3d-point-cloud-processing-clustering/https://s1s2.ai/2026/04/08/3d-perception-a-lidar-camera-pipeline-part-3-3d-point-cloud-processing-clustering/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Wed, 08 Apr 2026 00:00:00 GMThttps://s1s2.ai/2026/03/31/3d-perception-a-lidar-camera-pipeline-part-2-the-ros-2-graph-perception-data-flow/https://s1s2.ai/2026/03/31/3d-perception-a-lidar-camera-pipeline-part-2-the-ros-2-graph-perception-data-flow/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Tue, 31 Mar 2026 00:00:00 GMThttps://s1s2.ai/2026/03/26/3d-perception-a-lidar-camera-pipeline-part-1-setting-up-your-development-environment/https://s1s2.ai/2026/03/26/3d-perception-a-lidar-camera-pipeline-part-1-setting-up-your-development-environment/This article is part of a hands-on series building a 3D Perception Pipeline with ROS 2 Jazzy. In modern robotics, seeing isn’t enough —...Thu, 26 Mar 2026 00:00:00 GMThttps://s1s2.ai/2026/01/21/robot-vision-rosifying-a-yolo-pipeline/https://s1s2.ai/2026/01/21/robot-vision-rosifying-a-yolo-pipeline/Build a full robot vision pipeline: turn your YOLO script into a real ROS 2 perception node.Wed, 21 Jan 2026 00:00:00 GMT