Smart battery swapping station compatible with multiple battery packs, control method thereof, device, and medium

Patent metadata

Field	Value
Publication number	US-12370919-B2
Application number	US-202519060553-A
Country	US
Kind code	B2
Filing date	Feb 21, 2025
Priority date	Sep 23, 2024
Publication date	Jul 29, 2025
Grant date	Jul 29, 2025

Abstract

The invention relates to the technical field of battery swapping for electric vehicles. A control method of a smart battery swapping station includes: acquiring vehicle information of a vehicle requiring battery swapping, and retrieving a model type of a corresponding battery pack locking and unlocking hole and specific parameters from a database; when the vehicle arrives at a battery swapping operation platform, capturing an image of a vehicle chassis to determine an initial pose of the vehicle chassis; processing the initial pose to extract center coordinates of the locking and unlocking hole and a normal vector direction of the locking and unlocking hole; comparing the center coordinates of the locking and unlocking hole and the normal vector direction of the locking and unlocking hole with the retrieved model type of battery pack locking and unlocking hole and specific parameters in the database to determine if the information is consistent.

First claim

The invention claimed is:

1. A control method of a smart battery swapping station compatible with multiple battery packs, comprising the following steps:
acquiring vehicle information of a vehicle requiring battery swapping, and retrieving a model type of a corresponding battery pack locking and unlocking hole and specific parameters from a database;
when the vehicle arrives at a battery swapping operation platform, capturing an image of a vehicle chassis to determine an initial pose of the vehicle chassis;
processing the initial pose to extract center coordinates of the locking and unlocking hole and a normal vector direction of the locking and unlocking hole;
comparing the center coordinates of the locking and unlocking hole and the normal vector direction of the locking and unlocking hole with the retrieved model type of battery pack locking and unlocking hole and specific parameters in the database to determine if the information is consistent;
if inconsistent, prompting a user to leave the battery swapping operation platform; if consistent, selecting, by a battery swapping automated guided vehicle (AGV), a corresponding locking and unlocking operation method, and retrieving a battery from a storage for battery replacement; and
once the battery swapping is complete, generating a battery swapping report and allowing the user to leave,
wherein selecting, by a battery swapping AGV, a corresponding locking and unlocking operation method, and retrieving a battery from a storage for battery replacement comprises:
implementing space relocation design to allow the battery swapping AGV to switch between different locking and unlocking methods;
utilizing a height switching mechanism to automatically adjust the height of a locking and unlocking mechanism based on height differences of battery packs for different vehicle models; and
during the locking and unlocking process, making fine adjustments through adaptive floating based on changes in the position of the battery pack,
wherein selecting, by a battery swapping AGV, a corresponding locking and unlocking operation method, and retrieving a battery from a storage for battery replacement further comprises:
performing battery replacement through a dual AGV coordinated control method, specifically comprising:
connecting independently controlled battery swapping AGVs within the battery swapping station via a communication protocol;
performing time synchronization on the battery swapping AGVs: using the network time protocol (NTP) to ensure consistent timestamps across all AGVs and station control systems;
performing task planning based on task requirements, the current status of the AGVs, the order of vehicle arrivals, and the type of battery pack, and dynamically assigning tasks to each AGV; and
utilizing a scheduling algorithm to designate one AGV for disassembly and another AGV for assembly for the vehicle requiring battery swapping,
wherein performing battery replacement through a dual AGV coordinated control method further comprises:
autonomous fault monitoring and fault-tolerant control: when one AGV fails, the other AGV switches to a standalone mode to continue the battery swapping task and sends fault information, and
wherein performing battery replacement through a dual AGV coordinated control method further comprises:
using a Leader-Follower strategy to allow the two AGVs to maintain spacing and angular deviation within a set range, using a distance sensor and an angle sensor to monitor and adjust the spacing in real time and continuously detect the relative distance between the two AGVs, and employing a spacing control algorithm to ensure that the two AGVs maintain a safe distance while traveling on a single track.

2. The control method of a smart battery swapping station compatible with multiple battery packs according to claim 1 , wherein acquiring vehicle information of a vehicle requiring battery swapping comprises:
setting up a camera at an entrance of the battery swapping station to capture images of vehicles entering the battery swapping station;
performing size adjustment and color space conversion on the images;
using the Canny edge detection algorithm to extract vehicle contours, employing the Hough transform method to detect geometric shapes within the vehicle contours, and acquiring point cloud data of the vehicle;
utilizing the Euclidean clustering algorithm to group different objects in the point cloud data, extracting key feature points of the vehicle contours from the clustered point cloud, and merging image features with point cloud features to form a complete feature vector; and
loading a pre-trained deep learning model, inputting the feature vector into the deep learning model, and outputting a classification result of the vehicle.

3. The control method of a smart battery swapping station compatible with multiple battery packs according to claim 1 , wherein capturing an image of a vehicle chassis to determine an initial pose of the vehicle chassis comprises:
capturing a raw image of the vehicle chassis using a visual sensor, and cropping and resizing the captured raw image to a specified dimension;
utilizing a deep learning model for feature extraction, the deep learning model downsampling the input image, predicting a position of a center of a target bounding box and an offset thereof, and generating a feature map;
based on the feature map, predicting the position of the center of the target bounding box, the offset thereof, as well as the width and height of the target bounding box;
acquiring intrinsic and extrinsic parameters of a camera, converting predicted keypoint positions in the feature map to spatial coordinates of the raw image through warp Affine transformation, and generating point cloud data of the vehicle chassis based on a three-dimensional space reconstruction result;
applying voxel filtering to the point cloud data to remove noise points;
performing Euclidean clustering on the filtered point cloud, and using principal component analysis (PCA) to analyze the point cloud data to identify a primary direction of the point cloud data and estimate an approximate orientation of the vehicle chassis;
using the random sample consensus (RANSAC) algorithm to fit the point cloud data to exclude outliers, and obtaining a planar model of the vehicle chassis; and
determining the initial pose of the vehicle chassis based on PCA analysis and RANSAC fitting results.

4. The control method of a smart battery swapping station compatible with multiple battery packs according to claim 1 , wherein processing the initial pose to extract center coordinates of the locking and unlocking hole and a normal vector direction of the locking and unlocking hole comprises:
performing denoising and contrast enhancement preprocessing on the initial pose, and converting an original RGB image to an HSV (hue, saturation, value) color space;
setting thresholds in the HSV color space for segmentation, and using the Canny edge detection algorithm to identify edges in the image;
applying morphological opening and closing operations to remove small interference spots or connect broken edges, using the Hough circle transform to detect circular contours in the image, and initially locating the locking and unlocking hole;
setting detection parameters to obtain the center coordinates and radius of each detected circle, and fitting a planar model using the least squares method for each initially located locking and unlocking hole;
selecting a region around a center of the initially located locking and unlocking hole as the region of interest (ROI) within the planar model, using the least squares method to fit a plane equation of a point set within the region, and adjusting a position of the center based on a plane fitting result;
extracting a normal vector f

Assignees

• State Grid Jiangsu Electric Power Co Res Inst
• Nari Technology Co Ltd
• Nio Co Ltd
• State Grid Jiangsu Electric Power Co Ltd

Inventors

• Yuan Xiaodong
• Gan Haiqing
• Han Huachun
• Man Jin
• Li Qun
• Liu Yu
• Wang Mingshen
• Chen Liangliang
• Zhang Ning
• Wang Lihui
• Ruan Wenjun
• Miao Huiyu
• Pan Yi
• Yang Fengkun
• Zeng Fei
• Lv Shukang
• Dou Sheng

Classifications

• Y02T10/7072

  Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors · CPC title

• Y02T10/70

  Energy storage systems for electromobility, e.g. batteries · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G06T2207/20061

  Hough transform · CPC title

• G06V2201/06

  Recognition of objects for industrial automation · CPC title