Autonomous Machine Lane Control via Virtual Footprint Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous machines face challenges in maintaining lane boundaries due to position uncertainty, which can lead to potential collisions and operational inefficiencies in environments like mines where precise control is critical.
Innovation Solution
A control system that includes a positioning unit and navigation unit, which calculates a virtual two-dimensional footprint based on the machine's actual footprint and uncertainty value, simulates movement along a lane, and adjusts speed and direction to maintain margins within predetermined boundaries, ensuring the machine stays within lane boundaries even during position uncertainty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the autonomous machine operates with position uncertainty, then the machine can function in challenging environments, but the machine may violate lane boundaries and cause safety issues
Solution Approach 1:
The system performs preliminary actions by calculating a virtual footprint that anticipates potential position deviations before they occur. The virtual footprint is computed in advance based on the uncertainty value, and the machine control system uses this pre-calculated footprint to proactively adjust travel parameters, rather than reacting after a boundary violation occurs.
Solution Approach 2:
The system applies beforehand cushioning by creating a virtual footprint that extends beyond the actual machine boundaries in the direction of uncertainty. This virtual cushion zone allows the control system to plan conservative trajectories that account for potential positioning errors, effectively cushioning against future boundary violations.
2Reliability
If the machine control system uses a virtual footprint to account for position uncertainty, then lane boundary compliance improves, but computational complexity increases
Solution Approach 1:
The system creates a simplified copy of the machine's physical footprint in virtual space. This virtual footprint copy is manipulated computationally to represent potential position deviations, allowing the system to perform safety assessments on the copy rather than requiring complex real-time simulations of the actual machine's uncertain position.
Solution Approach 2:
The system transitions from three-dimensional spatial reasoning to two-dimensional footprint analysis. By projecting the machine's position uncertainty into a 2D virtual footprint representation, the computational problem becomes more tractable while still capturing the essential safety constraints for lane boundary compliance.
3Reliability
If the machine adjusts speed and direction frequently to maintain lane boundaries, then safety improves, but productivity decreases
Solution Approach 1:
The system applies partial action by adjusting travel parameters only to the extent necessary to maintain lane boundary compliance. Rather than making excessive conservative adjustments that would significantly reduce productivity, the system calculates the minimum required parameter changes based on the virtual footprint analysis, achieving adequate safety with minimal impact on travel efficiency.
Data Source
AI summary
An autonomous machine control system includes a positioning unit measuring position and orientation, and a navigation unit storing a route plan including an intended travel path along a lane. The lane has a width defined by a left-hand boundary and a right-hand boundary. The navigation unit receives an uncertainty value associated with the position or orientation, and creates a virtual two-dimensional footprint based on an actual machine footprint and the uncertainty value. The navigation unit also simulates movement of the virtual footprint along the intended travel path, calculates a left-hand margin value defined by the virtual footprint and the left-hand boundary, and calculates a right-hand margin value defined by the virtual footprint and the right-hand boundary. The margin values are compared to a predetermined value, and speed or travel direction of the machine is controlled if either of the margin values is below the predetermined value.