Mobile Machine Re-Routing Using Deep Learning Field Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing route planning systems for mobile machines, such as agricultural, construction, and forestry machines, do not adequately consider complex interactions between terrain, ground conditions, soil type, weather, and machinery capabilities, leading to suboptimal paths and increased operational costs.
Innovation Solution
The use of machine learning, specifically deep learning models based on Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, or Transformer-based models, to update and re-route mobile machines in real-time, taking into account various factors such as terrain, weather, and machine capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual path planning or simple heuristic algorithms are used, then the system complexity is low and ease of operation is maintained, but the route planning quality deteriorates leading to increased fuel consumption and reduced productivity
Solution Approach 1:
The patent replaces manual path planning and simple heuristic algorithms with a machine learning-based computing system that automatically analyzes complex interactions between terrain, ground conditions, soil type, weather, and machinery capabilities. This substitution transitions from mechanical/manual operations to an automated intelligent system, resolving the contradiction by accepting increased system complexity to achieve significantly improved route planning quality and operational efficiency
Solution Approach 2:
The patent introduces a computing system with machine learning models as an intermediary between the operator and the path planning process. This intermediary automatically processes multiple complex factors (terrain, weather, soil conditions, machine capabilities) that would be difficult for a human operator to consider simultaneously, thereby improving route planning quality without requiring the operator to directly manage the complexity of these interactions
2Use of energy by moving object
If complex interactions between terrain, ground conditions, soil type, weather, and machinery capabilities are not considered, then the ease of operation is maintained, but the route planning quality deteriorates resulting in increased fuel consumption
Solution Approach 1:
The patent implements a feedback mechanism where the computing system continuously monitors actual route performance and compares it with planned routes, using this information to refine and update route plans in real-time. This feedback loop enables the system to adapt to changing conditions and optimize fuel consumption dynamically, resolving the contradiction by automating the complex analysis of multiple factors that would otherwise require extensive manual operational complexity
3Productivity
If real-time route updates and machine control are implemented, then the productivity is improved, but the device complexity and extent of automation increase
Solution Approach 1:
The patent implements dynamic route planning where the computing system continuously updates route plans in real-time based on changing field conditions, machine status, and operational progress. This dynamic approach allows the system to adapt and optimize routes during operation, significantly improving productivity by responding to actual conditions rather than following static pre-planned routes, while the automation handles the complexity of real-time decision-making
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
Embodiments include technologies that use machine learning to update a route plan to reroute a mobile machine in a field. In some examples, a method includes using a mobile machine (110) to perform work in a field and recording travelled path information (step 402). The recorded path information (118) includes information about one or more paths followed by the machine (110) when performing the work in the field. The method also includes using one or more computing devices on the machine (e.g., see computing system 102 or 200) to input the path information into a trained deep learning model (107b) and to receive new machine travel paths (120) from the trained model (step 404). Also, the method includes using the computing device(s) on the mobile machine to control the machine to follow the new travel paths (120) generated by the trained model (step 406).