Data and Information Lead to Better Design

I admire the capability of data and information, especially in the AEC field, utilizing computational tool will lead us towards better designs and future.

Since the Rhino inside Revit plugin come up, I have been exploring to integrate a work-flow which can automate part of the works for architects, and here is a clip showing an example. And in this case, the basic logic is to do most of the geometry and design exploration works in Grasshopper/Rhino, push building informations to Revit and categorize them, and then try to do all the documentation from Revit using Dynamo.It's a very simple demonstration, and most of the scripts are really easy to develop, but by such a simple workflow, we can already increase lots of productivity.

Pokémon Raining game!

Last year, the brainiacs at Tehran University unleashed the Euglena plugin, a mind-blowing concoction of Object Detection and Face Recognition riding the Rhino 8 wave. （ Thanks Mahdiyar Esmailbeigi for the introduction!!🫶🫶）
Oddly enough, it played hard to get and refused to install on my device. 🤔

But hey, curiosity got the better of me, especially when it came to Hand Detection. So, I rolled up my sleeves and decided to craft my own digital magic. ✨ After a bit of detective work, I cracked the code – turns out, you can shuttle Python data to Grasshopper using UDP. 🕵️‍♂️

Now, Hand Detection is undeniably cool, but I found myself scratching my head, wondering, "What the heck do I do with this?" 🤷‍♂️ So, in a stroke of brilliance (or madness), I birthed a whimsical masterpiece – behold, the Pokémon Raining game! 🌧️🎮

🤌🤌 The second hand-tracking video is here! This time, I’ve built upon the same hand-tracking code as before, but I’ve added three exciting new features:

a. Implementing click actions by tracking finger taps.

b. Using Kangaroo2 to simulate dynamic gravity, physics, and collision scenes.

c. Creating a data loop by feeding the newly generated output back into Grasshopper. You’ll see new balls being added to the scene, which will still interact with other balls within the same gravity collision box. This was a question raised by a viewer and, in my opinion, the most challenging part of the video.

Overall, it’s an incredibly fun and fascinating challenge! 🤓🤓

Enjoy! 😁

Augmented Reality! This is hands down the coolest thing I've ever cooked up: Augmented Reality! 🚀🚀
You just need a iPad, and with the magic of the Aruco Library from OpenCV and some OpenGL wizardry, I managed to slap a 3D model onto an Aruco Marker, and the level of detail is mind-blowing! You can even give it a spin by scaling or rotating. I know it's not groundbreaking, but it's so darn fascinating that I couldn't wait to spill the beans! 😎😎

Recently, I encountered a task that inspired me to develop a specific workflow to address it. The challenge involved a manufacturer's library, brimming with an extensive array of families. However, the sheer volume made updating or managing any particular family either a Herculean task or akin to finding a needle in a haystack. 📚📚
In response to this daunting task, I crafted a workflow divided into three key steps:
a. 📊Processing the original data and creating an Excel document to catalog them. Each family was assigned a unique parameter value, enabling me to later track each family with ease. 📊
b. 🛠️Using 100 families as a pilot, I added a shared parameter to each, setting the value according to the Excel documentation. This method facilitated easy identification and management. 🛠️
c.🔍 Developing a simple 'search engine' to locate a family by its corresponding parameter value, streamlining the search process significantly. 🔍
This workflow proved highly effective for this particular challenge, though it's important to tread carefully. For instance, each family may contain multiple types, potentially complicating the data structure and jeopardizing the automation process. In other words, a thorough review of the original data is crucial. 🚦🚦

🚗 🚗Check out this cool project: Garage Path Planning! It all started when a fellow internet user asked me a question about monitoring a garage and figuring out how many parking spaces were available. Seemed like a straightforward task, but I decided to take it up a notch.
🛠️ I crafted a script not only to monitor but also to plan the optimal parking path for you. As I delved into it, I realized this concept isn't new—it's a hot topic in the world of robotics and extends seamlessly into the realm of computer vision. How exciting is that!
🧠 While tackling this challenge in the Grasshopper and ghpython environment wasn't a walk in the park, I gleaned some valuable insights into techniques like RRT (Rapidly Exploring Random Tree), Greedy Algorithm for finding the shortest path, and Dijkstra algorithm for evaluating the path and cost.
🏢 It's mind-blowing how these computational methods can elevate architecture, especially when we're the creators of the raw data, like the floor plan of a garage in this case. The possibilities are truly exhilarating!

Can we somehow transform a regular picture into data that CAD programs can understand? 🤔
Well, I delved into it and discovered why it's such a puzzle. 😩 😵
The real brain-buster is figuring out the colors and sorting them out. That's where the magic of machine learning kicks in, especially thanks to LunchBoxML and its K-Means Clustering. With that nifty tool, I managed to whip up a script that transforms any image into sleek curves.
😎 😎 Picture this: turning textures into city plans or cool facade patterns. And hey, it's not just for serious stuff – imagine making your own coloring book! 💁‍♂️ 💁‍♀️

When it comes to modelling facade in Revit, it's sometimes can be very tricky.
On one hand the the Curtain Wall-Panel way can let you easily swap the Panels from you self-made family, which is really handy. However on the other hand, when you have to manually adjust each facade grid-UV lines, it's really painful.
Here I created a example, to demonstrate how we could generate the facade grid controlled by the floor plan grid, which allows you to adjust your floor plan grid and automatically have a new facade, with adjustable native family (Generic Adaptive).
The process shows how we could combine the power of geometry manipulation from Rhino-Grasshopper, and through the proper setup to pass the data to Revit and then do the further processing.

LPH4-5 This workflow shows a process of data analysis and processing. The algorithm was designed to read and process the original data from the client, extract the useful information and most importantly, automatically structuring and documenting them into a certain format (in this case, a excel file) for better communication and post application.

LPH1-3 This algorithm was designed to quickly test different facade units, compare their overall visual appearance, and in the end help us better determine the directions of the development.

LPH1-3 - Grid Optimization. Through the optimization algorithm, minimize the differences among the facade units, find out the minimum types (sizes ) which are needed to build the facade, and maximize the productivity.

LPH2-3 - Real time radiation analysis shows how a high-rise building can be affected by the regional climate as well as the surroundings in Manhattan area.

LPH1-3 - Master plan study shows through changing parameters, we can provide quick generation and comparison of different variants.

Here we have a follow-up clip that demonstrates how we can quickly explore Geometry and Facade options and synchronize the data to Revit for further processing.
In the initial stages, we can begin by creating rooms, areas, schedules, and more. In this particular case, I utilized Dynamo to swiftly place doors on each floor in a matter of seconds.

LPH1-4 The Wiesbaden city passage is an important project in Wiesbaden which attracts lots of public spotlights. The site locates at a very central position of wiesbaden, lays between the Schwalbacher straße and Kirchgasse, within the premises, there are existing buildings and public facilities, moreover, heritage building and an ancient time city wall, which bring more complexity to the project.

Luoyang art gallery is strongly embedded in local and global environment and culture. its dialogue with the surrounding, including aspects relating to the green areas and scenery of the province, adjacent urban landscape and axis on which the project itself is situated, are emphasized by its harmonious form and position of the entrances.

LPH1-5 Revit + Real time structural analysis, to help us better determine the form of the roof at the same time optimize the structure of the variation.

A clip showcasing my sketches works