Recent content in Home on Gilbert Tanner
Hilti × Trimble SLAM Challenge 2026 entry built for the 3D Vision course at ETH Zürich — a metric, gravity-aligned, floorplan-consistent trajectory and 3D map from indoor dual-fisheye IMU data, ranking 3/22 in localization and 7/62 in SLAM.
A comprehensive guide to setting up multiagent simulation environments for PX4 and ArduPilot with Gazebo, covering namespacing, sensor configuration, and external odometry.
Building a pipeline that converts spoken language into continuous sign language motions for a humanoid robot.
This project presents a multi-mission transformer robot capable of operating both as a ground vehicle and a quadrotor drone . The design integrates constant-velocity (CV) joints , slip rings , and a servo-driven lifting mechanism to enable seamless transformation between driving and flying modes. We also utilize the dual-use wheel-propeller system , where each wheel houses a brushless drone motor…
This project implements real-time gesture classification using the uMyo EMG sensors . The system is designed to be used for precision control applications, such as controlling a robotic arm or a drone. On that front we implemented two demos controlling a simulation drone via Ardupilot Gazebo simulation and a Ryze Tello drone in real life. The project was implemented as part of the Pervasive…
Extended an existing Deep Reinforcement Learning navigation framework to support the TurtleBot3 platform with realistic 2D LiDAR observations, and migrated the full system from ROS 1 to ROS 2. The project enables training DRL agents for goal-directed mobile robot navigation with obstacle avoidance in Gazebo, using LiDAR-based state representations and velocity-based control tailored to TurtleBot3…
Through a series of collaborative competitions, the Sapience initiative fosters innovation in search and rescue operations, enabling multiple drones to effectively navigate and map GPS-denied environments, detect and deliver aid to victims, and perform complex cooperative tasks.
This page contains a short collage of some of the hardware projects I worked on over the years, starting with school projects created when I was studying at HTL Mössingerstraße. Coil gun / Coil winder Start: 11.2017 Project members: Gilbert Tanner Gabriel Tanner Alexander Pichler Aaron Armbruster Project supervisor: Herwig Guggi Figure 2: Coil gun / Coil winder Self Driving RC car Start: 10.2018…
HASCY is a remote-controlled sled that drives on a rail that is mounted at the top of a tunnel. It is equipped with multiple sensors, including a thermal and optical PTZ camera, and therefore is able to provide visual information to the Asfinag operators as well as emergency services.
Using either the TFJS Task API or the TFLITE Web API you can now deploy Tensorflow Lite models on the web without even needing to convert them into Tensorflow.js format.
The TensorFlow Lite Model Maker library is a high-level library that simplifies the process of training a TensorFlow Lite model using a custom dataset. It uses transfer learning to reduce the amount of training data required and shorten the training time.
D2Go is a production-ready software system from FacebookResearch, which supports end-to-end model training and deployment for mobile platforms.
TensorFlow.js is a deep learning library providing you with the power to train and deploy your favorite deep learning models in the browser and Node.js.
With the recently released official Tensorflow 2 support for the Tensorflow Object Detection API, it's now possible to train your own custom object detection models with Tensorflow 2.
Learn how to use the Tensorflow Object Detection API with Tensorflow 2
The Arduino Nano 33 BLE Sense is an evolution of the traditional Arduino Nano, but featuring a lot more powerful processor, the nRF52840 from Nordic Semiconductors, a 32-bit ARM® Cortex™-M4 CPU running at 64 MHz.
Use TensorRT to run PyTorch models on the Jetson Nano.
Run Tensorflow model on the Jetson Nano by converting them into TensorRT format.
YOLO Object Detection on the Jetson Nano using TensorRT
The NVIDIA Jetson Nano Developer Kit is a small edge computer for AI development. The Jetson Nano Developer Kit packs a Quad-core ARM A57 CPU with a clock-rate of 1.43GHz and 4GB of low-power DDR4 Memory.
Train a custom yolo object detection model in PyTorch
Use and create YOLOV3 models with keras-yolo3.
Use YOLOv3 with OpenCV to detect objects in both images and videos.
Learn how to use YOLO for Object Detection.
Getting started with Mask R-CNN in Keras
Create a custom Mask R-CNN model with the Tensorflow Object Detection API.
Learn how to create a custom instance segmentation model using Detectron2.
The LoRaWAN® specification is a Low Power, Wide Area (LPWA) networking protocol designed to wirelessly connect battery operated 'things' to the internet in regional, national or global networks.
LoRa is a spread spectrum modulation technique derived from chirp spread spectrum (CSS) technology. LoRa allows for long-range, low power wireless communication, often applied in IoT (Internet of Things) applications.
Creating math animations in Python with Manim, a mathematical animation engine made by 3Blue1Brown
Use your Tensorflow Object Detection model on edge devices by converting them to Tensorflow Lite.
Deploy your Streamlit application using Heroku, a platform as a service (PaaS)
Interpret PyTorch models with Captum.
Detectron2 is Facebooks new vision library that allows us to easily us and create object detection, instance segmentation, keypoint detection and panoptic segmentation models. Learn how to use it for both inference and training.
Turn your data science scripts and projects into beautiful websites/dashboards using Streamlit.
One of the most popular languages today is C# which is used for many applications. To use the power of Machine Learning in C# Microsoft created a package called ML.NET which provides all the basic Machine Learning functionality.
Local model interpretation is a set of techniques aimed at answering questions like: Why did the model make this specific prediction? What effect did this specific feature value have on the prediction?
Global model interpretation is a set of techniques that helps us to answer questions like how does a model behave in general? What features drive predictions and what features are completely useless for your cause.
The Google Coral Edge TPU allows edge devices like the Raspberry Pi or other microcontrollers to exploit the power of artificial intelligence.
Regardless of what problem you are solving an interpretable model will always be preferred because both the end-user and your boss/co-workers can understand what your model is really doing.
Learn how to create your own object detector using the Tensorflow Object Detection API.
Get started visualizing data in Python using Matplotlib, Pandas and Seaborn
Learn the basics of Keras, a high-level library for creating neural networks running on Tensorflow.
Scrape data from Reddit using PRAW, the Python wrapper for the Reddit API.
Build a system that is able to recommend books to users depending on what books they have already read using the Keras deep learning library.
Generating text in the style of Sir Arthur Conan Doyle using a RNN
Gilbert Tanner (2021) Gilbert Tanner is a robotics researcher and currently is doing his Master in Robotics, Systems and Control at ETH Zürich . For his Bachelor he studied Robotics and Artificial Intelligence at the University of Klagenfurt where he also worked on multi-agent drone research . In high-school he studied Electronics and Computer Science at HTL Mössingerstraße . For his diploma…