SBIR Phase II: Cloud-based Acoustic Simulation Service

Impulsonic, Inc., Carrboro NC

Investigators

Abstract

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project results from the proposed technology touching many aspects of architectural design. The company's cloud-based acoustic simulation service (CBASS) is aimed at architects and acoustical consultants. There are over 105,000 architects, 25,000 architecture students, and 2,000 acoustical consultants in the US. Acoustic simulation is a $450 million market. CBASS will be used to understand a building's acoustics before it is built, reducing the time and money spent on fixing acoustic issues. CBASS will check for acoustic issues in a design and automatically suggest ways of fixing them, allowing architects to be more certain of their designs, and saving on remodeling costs by getting the acoustics right initially. CBASS will also speed up and simplify the process by which architects and acoustical consultants collaborate, leading to reduced costs and better acoustics. The company will offer low-cost subscriptions to students, to help train the next generation of architects to think about acoustics earlier in the design phase. Increased use of acoustic simulation will lead to better sounding spaces for everyone: classrooms that are more conducive to learning, hospitals that more conducive to healing, more enjoyable theaters, and quieter homes. This Small Business Innovation Research (SBIR) Phase II project aims to develop new ways of using cloud computing to understand and improve the acoustics of buildings. Reliably modeling the behavior of sound waves is a hard problem. Researchers have recently developed an algorithm called Adaptive Rectangular Decomposition (ARD), which can accurately model acoustics. To make it work on complex, real-world data, the company has developed algorithms for running ARD on multiple computers at the same time, in the cloud. In this project the company will develop new algorithms that use acoustic modeling and machine learning to automatically check for compliance with standard acoustic guidelines, identify acoustic issues, and suggest changes and improvements to a design that are likely to improve its acoustics. When running ARD in the cloud, individual computers must keep sending data to each other, slowing down the overall simulation. The company will develop improved ways of managing this data exchange, allowing much faster and more reliable acoustic simulation. To let teams of architects and acoustical consultants collaborate on a project, the company will also develop ways of allowing the same project data and analyses to be accessed and reviewed by multiple people at the same time.