Self-Learning Error Avoidance for Autonomous Floor Processing Robots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automatic tillage devices often encounter recurring error situations that cause them to stop or malfunction, leading to incomplete processing of surfaces, as they lack the ability to predict and prevent such errors proactively.
Innovation Solution
The tillage device is equipped with a self-learning control system that analyzes detected parameters using artificial intelligence methods, such as machine learning, to identify recurring patterns and error causes, allowing it to predict and prevent future errors by adjusting its operation or alerting the user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automatic navigation and self-location are implemented, then the tillage device can operate autonomously, but recurring error situations cause work process interruptions and incomplete processing
Solution Approach 1:
The control device performs preliminary analysis of detected parameters to identify recurring error patterns before they cause complete system failure. By detecting parameters such as position, speed, and operational status in advance, the system can predict potential errors and take preventive actions to maintain continuous operation.
Solution Approach 2:
The system continuously monitors detected parameters and feeds this information back to the control device, which analyzes patterns over time. This feedback loop enables the system to learn from recurring errors and adjust its operation to prevent future interruptions, thereby maintaining reliable autonomous operation.
2Productivity
If the tillage device continues operation despite errors, then productivity is maintained, but the same errors recur preventing complete processing
Solution Approach 1:
The control device analyzes detected parameters in advance to identify recurring error patterns before they interrupt the work process. By predicting potential errors based on historical parameter data, the system can take preventive measures to avoid interruptions and ensure complete processing.
Solution Approach 2:
The system dynamically adjusts its operation based on analyzed parameter patterns. When recurring error patterns are detected, the control device modifies operational parameters or navigation paths to avoid the error conditions, thereby maintaining continuous productivity without repeating the same mistakes.
3Reliability
If manual intervention is required to resolve errors, then the device can be freed from error situations, but autonomy is reduced and operation is interrupted
Solution Approach 1:
The control device automatically analyzes detected parameters to identify and resolve recurring error patterns without requiring manual intervention. The system independently adjusts its operation or navigation to prevent errors, maintaining full autonomy while ensuring reliable continuous operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a soil cultivation device (1) that moves automatically within an environment, comprising a drive unit (2), a soil cultivation unit (3, 4, 5), an obstacle detection unit (6, 7), a control unit (8) and a detection unit (9, 10, 11, 12, 16) for detecting device parameters and/or environmental parameters, wherein the control unit (8) is configured to determine, based on the detected parameters, a fault of the soil cultivation device (1) which prevents movement of the soil cultivation device (1) and/or cultivation of an area to be cultivated by the soil cultivation device (1) in such a way that the soil cultivation device (1) cannot automatically free itself from the fault situation.To prevent the repeated occurrence of the same error situations, it is proposed that the control device (8) be configured to analyze the parameters detected by the detection device (9, 10, 11, 12, 16) for the purpose of self-learning error avoidance with respect to recurring patterns (13) characterized by a repeatedly occurring combination of an error and at least one temporally preceding environmental and/or device parameter.