Plant Agents
This project aims to create a scalable solution for solving seemingly incomprehensible spatial design problems through simple comprehensible inputs. We have experimented with how an agent-based design approach can achieve this in the domains of planting, site planning, and land use planning.
Planting design, for example, is a challenging problem that often requires high levels of horticultural expertise. By creating tools that tap into large sets of existing plant data, we believe that we can help to make planting design and maintenance more sustainable. This project is entering a partnership with a larger group of national stakeholders and planting experts call PlantAgents.
By creating tools that tap into large sets of existing plant data, we believe that we can help to make planting design and maintenance more sustainable.
Development
This project strives to handle complexity through simplicity. Like an ant-colony, the agents each have very simple rules which result in more complex, emergent behavior.
While machine Learning is all the rage, this project eschews that trend and relies instead on agent-based modeling to solve complex design problems. Each little agent represents a part of the overall design. The user defines design objectives for each agent. These objectives drive the agents to each behave differently, ultimately resulting in a design solution.
Strategy
The agents have an incredibly strategic role to play in the future of design. As we do more and more with computation, we need to learn to leverage the tools in the right places to empower and educate designers.
