Object Classification Using Distance and Oscillation Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object classification methods in autonomous driving and robotics lack accuracy due to reliance on single data types, such as distance information alone, and may not effectively handle complex environments with human presence, where visible wavelengths can be hazardous.
Innovation Solution
A method utilizing both distance and oscillation information from electromagnetic signals, particularly in the 700 nm to 1,300 nm wavelength range, to classify objects, which includes using a machine learning system that processes these signals for improved accuracy without additional sensors or external information, enabling classification based on reflected radiation and phase measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only distance information is used for object classification, then the system is simple and cost-effective, but classification accuracy is insufficient
Solution Approach 1:
The patent combines distance information and oscillation information from electromagnetic signals into a unified classification system. The machine learning model processes both types of information simultaneously, merging previously separate data streams to achieve improved classification accuracy without requiring entirely separate sensor systems.
Solution Approach 2:
The electromagnetic signal processing system is designed to extract multiple types of information (distance and oscillation) from the same signal source. This multi-functional approach allows a single sensor system to provide diverse data types that improve classification while avoiding the need for multiple specialized sensors.
2Reliability
If visible wavelength electromagnetic signals are used, then object detection is effective, but safety is compromised in environments with human presence
Solution Approach 1:
The patent changes the wavelength parameter of the electromagnetic signals from the visible range to the 700 nm to 1,300 nm range (near-infrared). This parameter change maintains the effectiveness of object detection while eliminating the safety hazards associated with visible light exposure to humans, as the new wavelength range is invisible and non-harmful.
3Measurement precision
If multiple sensor types are added to improve classification accuracy, then detection precision improves, but system cost and complexity increase
Solution Approach 1:
The system extracts both distance and oscillation information from the same electromagnetic signal measurements, making the sensor system multi-functional. This approach achieves improved classification accuracy without increasing the quantity of sensors, as one sensor system performs multiple measurement functions simultaneously.
Solution Approach 2:
The electromagnetic signals themselves carry multiple types of information (distance and oscillation characteristics) that are extracted through signal processing. The system serves itself by deriving diverse data types from the same measurement source, eliminating the need for additional specialized sensors.
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
Enhances classification accuracy and safety by providing additional object characteristics, reducing reliance on external data, and allowing for autonomous operation in environments with humans, while using existing technologies like lidar and radar for cost-effective implementation.
Implementation Method 1
The distance information may be a distance between the receiver and/or transmitter device and the object
Implementation Method 2
information about a phase difference between transmitted and received radiation
Data Source
AI summary
A method for classifying an object having the following: receiving at least one item of distance information of an object based on a first electromagnetic signal transmitted by a transmitter device and a first electromagnetic signal received by a receiver device; receiving at least one item of oscillation information of the object based on a second electromagnetic signal transmitted by a transmitter device and a second electromagnetic signal received by a receiver device, which represents a solid oscillation of at least one subsection of the object; and classifying the object based on the received information.