SPECIAL PROJECTS

2020

Our team of Design Associates (Michael Williams and Jordan Domjahn) and BMW Group Interns (Epifanio Pereira, Tim Lim, Rebecca Wickham) worked under leadership of Dylan Sheppard (idealworks) to 3D model, texture, render and animate the entire idealworks iw.hub Autonomous Mobile Robot, right down to every nut and bolt! The renders produced by the Academy and BMW Group interns coincided with the late 2020 announcement and birth of idealworks - the newest fully owned BMW Group subsidiary. Earlier iterations of the iw.hub (called the BMW Smart Transport Robot - STR for short) were designed and made by a team within BMW – this team is now at the core of the idealworks company.

This project took roughly 5 months to complete and has been one of the biggest Special Projects that the Academy has worked on to date. Updated renders were completed in mid 2021, and it’s likely that the Academy staff will complete further modelling and renders for new iw.hub marketing material in the future.

Check out the idealworks website to see more of the work created in this project.

2020 - Present

BMW has partnered with NVIDIA to create digital twins of BMW factories (using NVIDIA technology) to digitise their factory and logistics planning efforts, now and into the future. The purpose of this is to train Artificial Intelligence (AI) systems in a 1:1 scale digital environment without disrupting production on the factory floor. The training data learnt in the digital environment is implemented to a fleet of Autonomous Mobile Robots (AMRs) in the real world. The benefits that digital twins provide include factory layout planning, fixing and optimising workflows in real time and mitigating any errors or risk, without disrupting the factory floor.

Central to training the AI systems is object recognition. In short, this means that engineers can ‘tell’ the robot whether an object is a type of vehicle, person, container, or dolly etc. and how it should interact with each object. The robot continues to develop its object recognition capability as it is trained within the digital twin which uses digital replicas (3D assets) of real-life objects within BMW factories. Once a fully autonomous robot can identify different objects, it can make its own decisions when navigating spaces and performing tasks.

QUT’s involvement in this project has led to the creation of a large 3D asset library for software engineers at BMW Group and idealworks. By creating realistic BMW factory 3D models, such as storage part containers, dollies, shelving and industrial vehicles, BMW Group can utilise these 3D assets to train AMRs for object recognition within the AI software that it uses. Object recognition is critical for AMRs as they are expected to be fully autonomous and perform actions based on real-time data received through various sensors.

Our Design Associates Michael Williams and Jordan Domjahn began on the 3D asset creation during their internships at BMW in Munich, then continued once they returned to Australia and began work at the Academy as Design Associates. They modelled and visualised well over 100 factory assets for BMW Group, which then increased to over 200 assets once new interns – Epifanio Pereira, Tim Lim, Rebecca Wickham – started working in Germany in 2020.

In 2021 the 200+ assets were consolidated into an online library that is accessible for BMW and idealworks staff in Germany, plus NVIDIA in the USA. All of the assets comply with industry standards and are contributing to automation, Artificial Intelligence, smart machines, smart vehicles and smart factory developments for BMW.

We continue to create additional models in-house at the Academy.

For more information, check out the links below:
https://www.youtube.com/watch?v=6-DaWgg4zF8
https://www.youtube.com/watch?v=F0huXhFfgw0
https://blogs.nvidia.com/blog/2021/04/13/nvidia-bmw-factory-future/https://www.youtube.com/watch?v=PrwewnI8j08

2021 – 2022

In the lead up to the launch event and announcement of our M8 Competition, we created a small number of digital assets and experiences that we could show guests during the event. The aim of these was to demonstrate the possibilities of the technology along with some of our combined skills and knowledge.

The first was a 3D model of the M8 Competition exterior, done in both high resolution (for renders and animations in future) and a more optimized version for the Virtual Reality and Augmented Reality parts.

For the Augmented Reality (AR) experience we created 3 different models; the engine and annotations, an animation of the wheel, and lastly an animation to visualise the exterior surface of the M8. Using a game engine the models were exported as an AR application which activate off a unique 2D image. This allows the user to easily scan the activation image, then view the 3D models on a tablet device which are overlayed on the car through Augmented Reality.

The hero experience for the launch event was the Virtual Reality (VR) scene. This immersive experience allows the player to go from an exact digital twin of the Beta Lab space with the M8 in front of them, to being inside a BMW factory where they can see iw.hub robots moving around. The factory scene references our previous Special Projects and ongoing work with idealworks, in addition to showing the huge scale of BMW manufacturing plants that are all over the world.

These models and scenes are also valuable for Research Associates at the Academy who are exploring robotics and human machine interface topics, or wanting to utilise emerging technologies like VR and AR in their studies.

2021 – Present

Our team of Design Associates contributed to the digitisation efforts of the BMW Hams Hall engine plant. Working with a previous Design Intern and BMW staff in the UK, plus idealworks in Germany, we supported with 3D modelling and scene building which will be used to train the iw.hub robots and their AI system.

In 2022 we commenced a new project for Hams Hall, after being invited to assist with the presentation of a future vision project. We are creating an animation which illustrates the 2030 and beyond future vision for Plant Hams Hall, detailing their production strategies, digitisation and sustainability efforts, plus integration of emerging and collaborative technology. We delivered Part 1 of this project in January 2023 - a medium resolution version of the animation. Part 2, a highly detailed and more realistic version, will be delivered in Q3 2023 and will be made public shortly after. The full animation will be published here once complete.

Official BMW Plant Hams Hall images provided by BMW Group.

2022 – Present

Since 2022 we have supported idealworks and TechOffice Munich on the SORDI project. SORDI (Synthetic Object Recognition Dataset for Industries) is a collaboration between BMW TechOffice Munich, idealworks, NVIDIA and Microsoft that enables software developers to utilise industry standard datasets to train computer vision models. These datasets are comprised of synthetic images of a library of 3D assets; many of which are authored by the QUT interns and Academy staff.

The Academy has been involved in a number of other projects for the SORDI project, including: website renders and landing page animation, 2022 SORDI Hackathon promotional animation, and two animations which illustrate the Green Physics AI part of SORDI.

You can view the website here - SORDI.ai

2022

Through our collaboration with the Brisbane BMW dealership and marketing team, we were tasked with creating some Augmented Reality experiences for the launch of the new i7 vehicle. We created the AR with the intention of highlighting some design features and technology within the new i7, which encourages people to explore the car further and interact with it. The AR app was available for guests at the Brisbane launch of the vehicle, and will now continue to be used in the showroom.

Early access to the i7 vehicle gave us the opportunity to photograph and 3D scan the car, which became the reference for the 3D models and animations in the AR experiences. With the help and expertise of Mo Faramazian, our Design Associate Intern at the time, we built, tested and exported the AR application.

Launch event photos provided by Brisbane BMW.

2023

While we have been working with the BMW Tech Office team for a few years, most of our efforts have been towards their SORDI initiative. Some of our work is separate to SORDI, which can be seen below.

This animation visualises a point cloud pipeline that Tech Office staff are exploring, whereby digital twins can be generated using Artificial Intelligence and point cloud data from 3D scans of existing factories. The purpose of the project is to ultimately accelerate the creation of digital twins. Artificial Intelligence is used for scene reconstruction in the digital twin, and to place assets throughout the environment. It does this by using existing object recognition methods and 6D pose estimation, which allows the AI to determine elements such as if the containers are filled or empty, the age of the asset and how the object is arranged.

2023

Shown here is our most recent Special Project for our good friends at idealworks, visualisations of their new 3D scanning robot.

As idealworks expands its portfolio of products and services, last year, we helped the company with the branding design and application of their new 3D scanning robot, in addition to a collection of renders to be used for presentations and communication purposes.

We worked through several concepts for the branding, selecting a few variations to present the idealworks team. The final design uses colour and angled lines to highlight some key inputs on the robot and add visual interest. For the renders, we 3D modeled most of the robot, dolly, and charging dock over the supplied engineering CAD data, plus the VLX scanner and factory environment. We applied realistic materials and textures, plus our idealworks graphics and branding design.