The Seminar “Green Typologies: Machine Learning (ML) in Tree-Building Competition” at gtla* is distinguished by its strong connections to cutting-edge research outcomes. Students are introduced to a revolutionary concept, the ecolope, which is proposed with the aim of mitigating the adverse impacts of architecture and urbanisation and breaking through the ongoing human-nature dichotomy. The ecolope, i.e., ecological building envelope, unites the indoors and outdoors as a three-dimensional transitional space to provide habitats for different types of inhabitants, including humans, plants, and animals.

In this seminar, we narrow down the topic with a focus on tree canopy growth as a component of Urban Green Infrastructure (UGI) and its application to vertical open space design. Students will apply the output of recent studies on tree canopy development and its interaction with the development of urban environments in the Research Training Group “Urban Green Infrastructure” (RTG-UGI).

Study sites will be selected in the inner city of Munich, where green and grey infrastructure compete for limited space to support different human needs. Mixed groups of architecture and landscape architecture students will work together to analyse various interactions between different elements in the 3D indoor-outdoor space. After modelling the competition between building volume and tree canopies, students are expected to propose an integrated design aimed at enhancing spatial quality over time, including a planting and growth management plan for the trees and feasible modifications to the buildings.

To support the timeline design process (e.g., in 10, 30, and 80 years), a machine-learning-based tool (TreeML) in Rhino for predicting the tree crown geometry in relation to surrounding buildings will be used. The use of the prediction tool, along with the parametric design process in the Rhino-Grasshopper environment, will be guided by experts from gtla* as well as the research group.

Students with experience or interest in some of the following skills are highly recommended to join:

-Computer-aided design (Rhino and grasshopper);

-Information modelling;

-Data-driven design;

-Computer-aided analysis and simulation;

-Multi-criteria optimisation and evaluation.

To register, you must participate in the online test on October 15, at 17:00.

Zoom Link:

https://tum-conf.zoom-x.de/j/62237775727?pwd=7iLyPynzB3b68HUNb3GvwCfzIOFdbj.1
Meeting ID: 622 3777 5727
Passcode: 784764

Teaching Team:

Prof. Ferdinand Ludwig

Hadi Yazdi, M.Sc.

Xi Chen, M.A.

Language:

English

Contact: 

sissy_xi.chen(at)tum.de

Location:

Munich, Room 1325 - 0503, Main Campus

Dates: 

Thursdays, 14:00 - 18:00

First date (online announcement & test): 

15.10.2024, 17:00

Kick-off: 

24.10.2024, 14:00

Modules:

Green Typologies

LV-Nr.: 0000005583 (Seminar)

Seminar 6 ECTS / 4 SWS

MA Landscape Architecture
MA Architecture
BA Landscape Architecture (6th semester onwards)

Registration for seminar on TUMonline