Construction Vehicle Path Prediction for Task Area Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surveying methods for construction sites require pre-surveying or driving the perimeter of the task area, which is time-consuming and delays the start of tasks, especially when boundaries need to be determined before commencing work.
Innovation Solution
A system and method using a vehicle equipped with sensors and a computer system to record and map paths, predict new paths based on the shape of initial paths, and automatically steer the vehicle to ascertain task area boundaries, allowing for immediate task initiation without pre-surveying, using a 'teach and repeat' process that records and replicates tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-surveying or driving the perimeter of the task area is performed to determine boundaries, then boundary determination accuracy is improved, but task start time is delayed
Solution Approach 1:
The system performs preliminary actions by having the vehicle drive along predicted paths and record positional data before final boundary determination. The vehicle collects boundary information during normal operation rather than requiring a separate pre-surveying step, thus preparing boundary data in advance without delaying task commencement.
Solution Approach 2:
The system creates a copy of the task area boundary information through digital recording of vehicle positions along multiple paths. Instead of physical surveying markers or traditional measurement methods, the boundary is represented as digital coordinate data that can be processed and stored, enabling rapid boundary determination without time-consuming field surveying.
2Measurement precision
If multiple paths are mapped and analyzed to determine boundaries, then boundary accuracy is improved, but system complexity increases
Solution Approach 1:
The boundary determination process is segmented into multiple discrete paths that the vehicle traverses. Each path provides a portion of the boundary information, and the complete boundary is constructed by combining data from these separate segments. This allows accurate boundary mapping through simple, repeatable vehicle trajectories rather than complex single-pass surveying.
Solution Approach 2:
The vehicle serves multiple functions: it performs both the construction task and the boundary surveying simultaneously. The same vehicle used for earthmoving or other construction work also collects boundary data through its equipped sensors and positioning system, eliminating the need for separate specialized surveying equipment and reducing overall system complexity.
3Productivity
If automatic steering along predicted paths is implemented, then productivity is improved, but automation complexity increases
Solution Approach 1:
The system performs self-service by using the vehicle's own recorded path data to generate predictions for subsequent paths. The vehicle learns from its own operational patterns and uses this information to autonomously determine optimal future paths without requiring external intervention or complex centralized control, thereby improving productivity through simple autonomous behavior.
Solution Approach 2:
The system implements feedback by using actual vehicle position data and path completion information to refine and adjust subsequent path predictions. The boundary determination and path planning continuously improve based on feedback from actual vehicle operation, enabling automatic steering that adapts to real conditions while maintaining manageable automation complexity through iterative refinement.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Construction vehicles using a teach and repeat system can ascertain boundaries of a task area while working, without knowing a perimeter of the task area before starting the task. The system maps a first path of the vehicle by recording a position of a vehicle moving from a first start position to a first end position. A second path is predicted, based on a shape of the first path, the second path having a second start position and a second end position. A third path is mapped from a third start position to a third end position. Boundaries of the task area are ascertained based on the first path, the first start position, the second start position, the third start position, the third path, the first end position, the second end position, and the third end position.