Autonomous Lattice Traversal Planning from Operator Path Deviations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack an efficient method to generate specific traversal plans based on actual operator data, leading to inefficiencies in task completion, such as plowing or seeding, due to deviations from general plans, which are not adequately accounted for in existing systems.
Innovation Solution
A method involving a computing device that receives a general plan and actual operator traversal data, determines subsets within a tolerance range, formulates a specific plan, and optimizes it to minimize distance, allowing for efficient comparison and input into autonomous vehicles for improved task execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If autonomous vehicles use general plans for traversal, then task completion is achieved, but efficiency deteriorates due to deviations from optimal paths
Solution Approach 1:
The system performs preliminary analysis of actual operator traversal data before generating the specific plan. By pre-processing operator data, identifying patterns, and calculating optimal paths in advance, the system creates a data-driven specific plan that anticipates real-world deviations. This preliminary action enables the autonomous vehicle to follow more efficient paths from the start, reducing energy loss during task execution.
Solution Approach 2:
The system transforms the general plan into a specific plan by changing key parameters based on actual operator data. This includes adjusting traversal paths, speeds, and timing parameters to reflect real-world optimal practices. By modifying these parameters empirically rather than relying on theoretical general plans, the system achieves better fuel efficiency and productivity.
2Stability of the object's composition
If autonomous vehicles strictly follow general plans, then path consistency is maintained, but adaptability deteriorates due to inability to account for actual operator optimizations
Solution Approach 1:
The system creates a specific plan that copies and adapts proven successful patterns from actual operator traversal data. Rather than strictly following a theoretical general plan, the system replicates real-world optimized behaviors while maintaining the overall structure and goals of the general plan. This copying approach ensures both consistency with proven methods and adaptability to actual conditions.
Solution Approach 2:
The system introduces dynamics into the plan by making it specific and data-driven rather than static and theoretical. The specific plan adapts its parameters based on actual operator performance, allowing the autonomous vehicle to dynamically adjust to real-world conditions while maintaining overall mission consistency. This dynamic approach balances path consistency with operational adaptability.
3Measurement precision
If actual operator traversal data is used directly, then real-world accuracy is improved, but data quality deteriorates due to excessive spatial and temporal deviations
Solution Approach 1:
The system extracts only the valuable elements from actual operator traversal data by filtering out excessive spatial and temporal deviations. It identifies and removes outliers, unnecessary detours, and inefficient timing variations while preserving the core optimized patterns. This extraction process produces a specific plan with improved precision that maintains real-world accuracy without the noise of excessive deviations.
Solution Approach 2:
The system applies local quality by selectively incorporating deviations from operator data only where they improve efficiency. Rather than uniformly accepting or rejecting all deviations, it evaluates each local segment of the traversal data and incorporates beneficial variations while excluding harmful ones. This selective approach optimizes both accuracy and deviation control in different regions of the traversal path.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A specific plan for traversing a lattice of a geographic region is generated and provided to an autonomous vehicle. A general plan that includes the lattice and a planned sequence of operator traversal data of the lattice is received. Actual operator traversal data used to traverse the geographical region is determined. A tolerance range that defines an allowable deviation from the general plan is determined. A first subset of the actual operator traversal data that is within the tolerance range and a second subset of the actual operator traversal data that is not within the tolerance range is determined. A specific plan for traversing the geographical region is formulated based on the first subset of the actual operator traversal data that is within the tolerance range. The specific plan is provided to the autonomous vehicle for controlling traversal of the geographical region by the autonomous vehicle.