Vehicle cabin cleaning method, apparatus, device, medium, and product

By combining robotic arms with visual language models and diffusion models for automated control, the problems of low cleaning efficiency and incomplete pollutant treatment in unmanned vehicles have been solved, achieving efficient and automated cleaning of the vehicle cabin.

CN122275813APending Publication Date: 2026-06-26ZHEJIANG GEELY HLDG GRP CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG GEELY HLDG GRP CO LTD
Filing Date
2026-05-20
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

The interior cleaning of driverless vehicles and shared mobility vehicles relies on traditional manual cleaning, resulting in low operational efficiency, high labor costs, and existing vehicle cleaning equipment cannot flexibly handle heterogeneous pollutants, resulting in poor cleaning quality.

Method used

By employing a robotic arm combined with a visual language model (VLM) and a diffusion model (DiT), and by acquiring image information through a camera, the system performs motion prediction and automated control to achieve automated cleaning of the vehicle cabin, including identifying the type of pollutants and replacing the appropriate cleaning tools.

Benefits of technology

It achieves an automated closed-loop system for cleaning the vehicle cabin, improving cleaning efficiency, reducing operating costs, shortening vehicle downtime, and effectively handling various types of pollutants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122275813A_ABST
    Figure CN122275813A_ABST
Patent Text Reader

Abstract

This application discloses a method, apparatus, device, medium, and product for cleaning the interior of a vehicle. The method is applied to an electronic device, which is communicatively connected to an interior cleaning system installed on the vehicle. The interior cleaning system includes a robotic arm. The method includes: acquiring first image information, which includes a first interior image captured by a first camera installed inside the vehicle's interior, and / or a second interior image captured by a second camera installed on the robotic arm; predicting actions based on the first image information and pre-acquired task instructions to obtain a first target action sequence; and controlling the robotic arm to clean the interior of the vehicle's interior according to the first target action sequence. The entire process requires no human intervention, improving the efficiency of vehicle interior cleaning.
Need to check novelty before this filing date? Find Prior Art