Self-learning Control Algorithm for Harvesting Machine Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for automatic control of agricultural harvesting machines are limited by inaccurate sensor detection, requiring operator intervention for fine-tuning and lacking adaptability to individual operator preferences, leading to suboptimal operating results and operator dissatisfaction.
Innovation Solution
A system that includes a sensor for detecting crop properties, a control device using an algorithm to generate control signals for actuators, and an interface for receiving corrective inputs from observers, allowing the algorithm to be modified based on these inputs, thereby adapting to operator preferences and learning over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If sensor-based automatic control is used to determine operating parameters, then automation extent is improved, but measurement precision is insufficient leading to suboptimal operating results
Solution Approach 1:
The system implements a feedback mechanism where the observer monitors the actual operating results and provides corrective inputs that are fed back to the control device. This closed-loop feedback allows the system to self-correct based on actual performance, compensating for insufficient sensor precision while maintaining high automation. The corrective inputs from the observer serve as feedback signals that adjust the control parameters to achieve optimal operating results.
Solution Approach 2:
The system enables self-service through the learning algorithm that automatically modifies control parameters based on accumulated corrective inputs from the observer. Over time, the system learns from operator corrections and autonomously adjusts operating parameters without requiring continuous manual intervention, thereby maintaining automation while improving measurement accuracy through experience.
2Device complexity
If fixed algorithms are used for automatic control, then device complexity is reduced, but adaptability to operator preferences is poor
Solution Approach 1:
The control algorithm transitions from a static, fixed design to a dynamic, adaptive system that evolves over time. The learning algorithm continuously modifies control parameters based on corrective inputs from the observer, allowing the system to adapt to changing operating conditions and operator preferences while maintaining relatively simple hardware architecture.
Solution Approach 2:
The system changes the parameters of the control algorithm itself based on accumulated learning from corrective inputs. By modifying the algorithm's internal parameters and control strategies based on observer feedback, the system achieves high adaptability to operator preferences without requiring complex hardware modifications or multiple pre-programmed algorithms.
3Adaptability or versatility
If operator intervention is required for fine-tuning, then adaptability is improved, but productivity is reduced due to continuous manual input
Solution Approach 1:
The system performs preliminary action by pre-adjusting control parameters based on sensor data and historical learning before manual intervention is needed. The learning algorithm proactively modifies operating parameters in anticipation of optimal performance, reducing the frequency and intensity of required operator interventions while maintaining high adaptability.
Solution Approach 2:
The system ensures continuity of useful action by maintaining automatic control with minimal interruptions for manual fine-tuning. The learning algorithm continuously processes corrective inputs and smoothly adjusts parameters, ensuring uninterrupted harvesting operations while progressively improving adaptability through accumulated learning, thereby preserving productivity.
Data Source
AI summary
A system for automatic control of an operating parameter of a crop transport or processing device of an agricultural harvesting machine includes a sensor configured to detect at least one property of a crop or a parameter affected by the operating property and a control device disposed in communication with the sensor to receive signals therefrom. The control device is operable by means of the signals from the sensor using an algorithm to determine successive control signals for control of an actuator. The control signals affect the operating parameter of the crop transport or processing device that interacts with the sensed crop. An operator interface is included for input of corrective inputs for the operating parameter. The control device is connected to the interface and is operable to override the control signals sent to the actuator by means of the received corrective inputs.