Gabriel Giangi

Tag: Reinforcement Learning

  • Intelligent Hexapod

    Intelligent Hexapod

    This NSERC-funded project pushed the boundaries of intelligent hexapod robotics, making use of 3D printing and ROS2.

  • Part 3: Building the Hexapod Simulation Environment

    Part 3: Building the Hexapod Simulation Environment

    In this third instalment of the intelligent hexapod series, I explored the integration of the physical robot into a simulation environment using ROS2 (Robot Operating System 2). I provided clear instructions on the setup process, focusing on ROS2 concepts such as nodes, topics, and messages that facilitate efficient communication between components. Additionally, I discussed the…

  • Towards AGI in Robotics: An Exploratory Approach to Self-Maintenance

    Towards AGI in Robotics: An Exploratory Approach to Self-Maintenance

    Building upon the quest for General Artificial Intelligence (AI) in robotics, this article proposes a novel architecture for a self-maintaining robot. Employing a triad of Perception, Cognition, and Action modules, the approach leverages machine learning techniques, reinforcement learning, and control theory, allowing the robot to monitor and manage its energy consumption, troubleshoot faults, and perform…