Exploring Innovative Projects by Owen Oliver: Design and Engineering Insights

Discover the diverse portfolio of Owen Oliver, a design engineering student and project manager. From 3D printing to hardware solutions, explore his experiences in the Technik competition and endurance racing, showcasing his skills in managing teams and creating essential components for high-performance race cars.

5/8/20241 min read

A 3D-rendered exploded view of a rectangular metallic object, with various layers and screws depicted as if floating apart. Each component appears precise and geometrical, showcasing an interconnected assembly of a product or part.

Exciting news in the world of robotics! Physical Intelligence has just open-sourced their new FAST model—a next-gen robotic foundation model designed to handle everyday tasks like folding laundry, cleaning tables, and even scooping coffee beans. Trained on an impressive 10,000 hours of example data, this model paves the way for robots to learn and adapt with unprecedented efficiency.

What’s especially cool is that the FAST variant leverages a state-of-the-art tokenizer for improved language following, making it easier than ever for developers and researchers to fine-tune the model for their own robotic platforms. This is a huge step forward in democratizing access to cutting-edge robotics technology, sparking innovation and collaboration across the community.

Kudos to the team at Physical Intelligence for pushing the boundaries of what’s possible. Whether you’re a robotics enthusiast, a researcher, or just passionate about AI and innovation, this open-source release is sure to inspire new ideas and projects.

Learn more about this breakthrough here: https://lnkd.in/exDAz6Ts