Agricultural Robot Crop Estimation Using Partial Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing agricultural robots face difficulties in accurately estimating crop types and postures due to challenges in obtaining and processing sensor data effectively.
Innovation Solution
The agricultural robot is equipped with an optical sensor and a crop estimator that utilizes deep learning to analyze sensor data, allowing for precise crop type and posture estimation by extracting relevant data and generating a surface condition model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is collected from various areas to identify crops, then crop identification accuracy should improve, but data processing complexity and time consumption increase
Solution Approach 1:
The patent segments the sensor data processing by dividing the collected data into multiple areas or regions of interest. The data extractor specifically designates and extracts partial data corresponding to designated crops from the overall sensor data, rather than processing all data uniformly. This segmentation reduces processing complexity while maintaining identification accuracy by focusing computational resources on relevant crop areas.
Solution Approach 2:
The patent applies extraction by using a data extractor component that isolates and extracts only the necessary partial data corresponding to specific crops from the complete sensor data set. This extraction process removes irrelevant information and focuses on crop-specific data, thereby reducing processing complexity and time while preserving the accuracy needed for reliable crop identification.
2Measurement precision
If comprehensive sensor data is collected to determine crop posture, then posture estimation accuracy improves, but data processing time increases
Solution Approach 1:
The patent segments posture determination by analyzing specific features and patterns within the sensor data rather than processing the entire data set. The estimator focuses on extracting posture-relevant information from designated crop areas, dividing the complex posture analysis into manageable feature extraction tasks that reduce processing time while maintaining accuracy.
Solution Approach 2:
The patent applies preliminary action by using the data extractor to pre-process and organize sensor data before posture estimation. The extraction of partial data corresponding to designated crops is performed beforehand, preparing the data in a format optimized for posture analysis. This preliminary extraction reduces the computational burden during the actual posture estimation process, decreasing processing time while preserving accuracy.
3Measurement precision
If deep learning is used to generate surface condition models, then crop estimation accuracy improves, but computational requirements and processing time increase
Solution Approach 1:
The patent applies extraction by using the data extractor to isolate only the necessary partial data corresponding to designated crops before applying deep learning models. This extraction removes irrelevant data that would otherwise consume computational resources, allowing the deep learning-based estimator to process only crop-relevant information. This reduces energy consumption while maintaining the high estimation accuracy provided by deep learning.
Solution Approach 2:
The patent applies partial action by using deep learning models only on the extracted partial data corresponding to designated crops, rather than applying them to the entire sensor data set. This partial application of computationally intensive deep learning techniques reduces overall energy consumption while maintaining crop estimation accuracy for the targeted crops.
Data Source
AI summary
An agricultural robot includes a traveling body, a working unit on the traveling body to do work relating to a crop, an optical sensor on the traveling body, and a crop estimator to perform estimation of the crop based on sensor data obtained by the optical sensor. The crop estimator includes a data extractor to designate the crop to be subjected to the estimation based on the sensor data and extract, as partial data, a portion of the sensor data that corresponds to the designated crop, and an estimator to perform the estimation of the crop based on a surface condition of the crop obtained from the partial data extracted by the data extractor.