Robotic EV Charging Port Detection Using Color Threshold Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric vehicle charging systems face challenges in accurately identifying the physical position of the EV charging portal, leading to inefficiencies and increased costs due to reliance on 2-D image processing, 3-D vision systems, and active fiducials, which are computationally expensive and susceptible to environmental conditions.
Innovation Solution
A plug connection system featuring a controllable robotic arm with an image capturing device that uses a pixel value threshold to filter images and determine the location of an optically distinctive object on the EV charging portal, enabling precise maneuvering of the charging plug, and optionally includes a light emitter for illumination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If 2-D image processing is used to identify the EV charging portal, then the system cost is reduced, but computational bandwidth and processing overhead increase significantly
Solution Approach 1:
The patent applies an optically distinctive object with specific color properties (e.g., retroreflective material, fluorescent coating, or colored adhesive) to the EV charging portal. The image capturing device captures images in specific color channels (e.g., blue channel for yellow objects, red channel for green objects) to enhance contrast and enable simple threshold-based pixel filtering, reducing computational requirements while maintaining identification accuracy
Solution Approach 2:
The patent changes the optical parameters of the charging portal by applying an optically distinctive object that reflects, absorbs, or emits light in specific wavelengths. This allows the use of simple 2-D image processing with pixel value thresholding instead of complex 3-D vision systems, significantly reducing computational bandwidth while maintaining identification precision
2Productivity
If 3-D vision systems are used to identify the EV charging portal, then computational efficiency is improved, but system cost and sensor complexity increase significantly
Solution Approach 1:
The patent uses an optically distinctive object that creates strong contrast in specific color channels, enabling efficient 2-D image processing with simple pixel thresholding. This approach achieves computational efficiency comparable to 3-D vision systems while using much less expensive 2-D cameras and processors
Solution Approach 2:
The patent employs inexpensive optically distinctive objects such as colored adhesives, stickers, or retroreflective materials that can be easily applied and replaced on the EV charging portal. These low-cost components enable efficient image processing without requiring expensive 3-D vision system hardware
3Measurement precision
If active fiducials with light sources are mounted on the EV, then the charging portal identification accuracy is improved, but vehicle cost and installation complexity increase
Solution Approach 1:
The patent uses passive optically distinctive objects with specific color properties that can be identified through color channel filtering and pixel thresholding. This approach maintains identification accuracy without requiring active light sources or complex electronic components on the vehicle
Solution Approach 2:
The optically distinctive object passively reflects or emits light according to its material properties, eliminating the need for active light sources, power connections, or electronic control systems. The object simply needs to be applied to the charging portal, greatly simplifying installation while maintaining identification precision
4Ease of manufacture
If QR codes are positioned on the EV charging portal, then the system cost is reduced, but reliability decreases due to damage and dirt accumulation
Solution Approach 1:
The patent uses an optically distinctive object with strong color contrast that can be reliably detected through color channel filtering and pixel thresholding. Unlike QR codes that require specific patterns and orientations, the colored object can be identified from various angles and distances, maintaining reliability while keeping costs low
Solution Approach 2:
The patent changes the identification approach from pattern recognition (QR codes) to color and optical property detection. This allows for more robust identification that is less susceptible to damage and dirt, as the optical properties of the colored object remain detectable even when partially obscured
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution provides a low-cost, computationally efficient method for identifying the EV charging portal, reducing processing overhead and enhancing the accuracy of automatic charging operations while minimizing environmental susceptibility.
Implementation Method 1
a light emitter for emitting illumination for the image capturing device
Implementation Method 2
an image capturing device configured to capture one or more images
Data Source
AI summary
A plug connection system that autonomously charges an electric vehicle (EV) is provided. The method includes: adjusting configuration settings of an image capturing device; capturing an image using the image capturing device, wherein a portion of the image comprises an optically distinctive object associated with a charging portal of the EV, and wherein the image comprises a plurality of pixels; applying a pixel value threshold associated with a pixel metric to the image to filter the plurality of pixels of the image to a first subset of pixels; determining, based on filtering the plurality of pixels to the first subset of pixels, a location of the optically distinctive object associated with the charging portal of the EV within the image; and providing information to maneuver a robotic arm to a physical position associated with the location of the optically distinctive object within the image.