The speed at which new technologies arrive on the market is nothing compared to the speed at which talented researchers find creative means to use them, to form them, even to transform them into things that we cannot live. One of these researchers is MIT Mad Fellow Alexander Htet KyawA graduate student who was pursuing two master’s degrees in architectural studies in calculation and electrical engineering and computer science.
Kyaw takes technologies such as artificial intelligence, augmented reality and robotics, and combines them with gesture, speech and recognition of objects to create human-AI work flows that have the potential to interact with our built environment, to change the way we buy, to design complex structures and do physical things.
One of his latest innovations is the Conservator AI, for which he and his MIT graduate student partners won the first prize – $ 26,000 in OPENAI products and species – during AI construction of the MIT conference: Generative Voice AI, a one -week hackathon at MIT with final presentations held last fall in New York. Working with Kyaw were Richa Gupta (architecture) and Bradley Bunch, Nidhish Sagar and Michael Won – All of the MIT Department of Electrical Engineering and Computer Science (EECS).
The AI curator is designed to rationalize online furniture purchases by providing contextual products recommendations using AI and AR. The platform uses AR to take the dimensions of a room with existing windows, doors and furniture. Users can then speak to the software to describe which new furniture they want, and the system will use an AI model in visual language to search and display various options that correspond to both the user’s prompts and the visual characteristics of the part.
“Buyers can choose from the suggested options, visualize AR products and use natural language to request research changes, which makes the furniture selection process more intuitive, efficient and personalized,” explains Kyaw. “The problem we are trying to solve is that most people do not know where to start when providing a room, so we have developed a conservative AI to provide intelligent contextual recommendations according to what your room looks like.” Although the Conservatives have been developed for furniture purchases, it could be extended for use on other markets.
Another example of Kyaw’s work is the estimate, a product that he and three other graduate students created during the MIT Sloan Product Tech Conference hackathon in March 2024. The objective of this competition was to help small businesses; Kyaw and the team decided to base their work on a painting company in Cambridge which employs 10 people. The estimate uses AR and an object recognition AI technology to take the exact measures of a part and generate a detailed estimate of costs for renovation and / or paint work. It also uses a generative AI to display images of the part or parts as they can look like after painting or renovation, and generates an invoice once the project is finished.
The team won this hackathon and $ 5,000 in cash. The Kyaw teammates were Guillaume Allegre, May Khine, and Anna Mathy, who all graduated from the MIT in 2024 with master’s degrees in commercial analysis.
In April, Kyaw will give a TEDX conference to his Alma Mater, at Cornell University, in which he will describe the curator AI, the estimate and other projects that use AI, AR and robotics to design and build things.
One of these projects is Unglog, for which Kyaw connected AR to the recognition of gestures to build software that takes the inputs of the fingertips on the surface of a material, or even in the air, to map the dimensions of the construction components. This is how UNLOG – an imposing art sculpture made from ash logs that are on the Cornell campus – appeared.
Recognition of gestures for mixed reality based on feedback and robotic manufacturing of the UNLOG tower
Video: Alexander Htet Kyaw
Unlog represents the possibility that the structures can be built directly from an entire newspaper, rather than making a log trip in a wood mill to be transformed into planks or two by four, then shipped into a wholesaler or detailing. It is a good representation of Kyaw’s desire to use building materials in a more lasting way. An article on this work, “Gesture recognition for the manufacture of mixed reality based on feedback A case study of the UNLOG tower“Was published by Kyaw, Leslie Lok, Lawson Spencer and Sasa Zivkovic in the acts of the 5th International Conference on Computer Design and Robotics Manufacturing, January 2024.
Another system developed by Kyaw incorporates physical simulation, gestures recognition and RA to design active bending structures built with bamboo posts. Gesture recognition allows users to handle digital bamboo modules in the RA, and physical simulation is integrated to visualize how bamboo folds and where bamboo posts in a way that create a stable structure. This work appeared in the acts of the 41st education and research in computer -assisted architectural design in Europe, August 2023, as “Active fence in mixed reality based on physics: the design and manufacture of a reconfigurable modular bamboo system. “”
Kyaw presented a similar idea using bamboo modules to create structures deployable last year in Mitdesignx, a MIT MID program which selects promising startups and provides coaching and financing to launch them. Kyaw has since founded Hairstyle To build prefabricated and modular bamboo shelters and community spaces for refugees and people displaced in Myanmar, its country of origin.
“Where I grew up, in Myanmar, I saw a lot of daily effects of climate change and extreme poverty,” explains Kyaw. “There is a huge refugee crisis in the country, and I want to think about how I can contribute to my community.”
His work with Bendshelters was recognized by the Sandbox MIT, the Social Innovation Challenge PKG and the Amazon Robotics Prize for Social Good.
MIT, Kyaw collaborates with Professor Neil Gershenfeld, director of the Center for Bits and Atoms, and doctoral student Miana Smith to use speech recognition, 3D generator and robotic weapons to create a workflow that can build objects in an accessible, on demand and sustainable objects. Kyaw holds a baccalaureate in Cornell architecture and computer science. Last year, he received a SJA scholarship from The Steve Jobs Archive, which finances projects at the intersection of technology and arts.
“I like to explore different types of technologies to design and do things,” says Kyaw. “Being part of Mad made me think about how all my work connects and helped to clarify my intentions. My vision of research is to design and develop systems and products that allow natural interactions between humans, machines and the world around us. ”