Press Release

[Solution Review] AgileSoDA’s ‘TwinCar’ provides automatic calculation of AI-based vehicle repair estimates 20/11/29

The AI ‘TwinCar(formerly NakedCar)’ developed by AgileSoDA recognizes the damaged part of the vehicle and measures its severity. TwinCar’s core algorithm has been verified its effectiveness and introduced in the ‘AOS Alpha’ project promoted by the Korea Insurance Development Institute. TwinCar recognizes the damaged parts of a vehicle by classifying only the pictures that show damaged parts. It is a solution that determines damage’s type and location and suggests relevant necessary repair. In 2018, it was firstly serviced by H Insurance’s AI system to automatically estimate repair cost.
AgileSoDA’s TwinCar is an AI application similar to the way a loss adjuster conducts an audit. It is designed to follow a unique process to predict the damage depth and repair cost by creating a rule that synthesizes the overall results based on various recognition and discrimination models for automatic repair cost estimate. Also, since each model of TwinCar is pre-trained through various images, it is very suitable for image-based transfer learning. Thus, by using images, it can be quickly applied to problems such as defect detection of PV panel, black box image analysis, facility crack safety management, and facility defect prediction/quality management.

서울시 강남구 선릉로 525 인포스톰빌딩 2-3층
| 대표번호(02-558-8300) | contact@agilesoda.ai
Copyright © AgileSoDA .