Two-Stage Planning for Autonomous Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current route planning systems for autonomous vehicles face challenges in efficiently navigating through uncharted environments due to limitations in sensor fidelity, particularly in the far field where measurements are less precise, leading to uncertainty and potential hazards.
Innovation Solution
A two-stage planning method that distinguishes between a near field with high fidelity measurements and a far field with low fidelity measurements, using sensor data to evaluate candidate plans, determine viability, and select a composite plan with the highest flexibility score, ensuring robust and resilient navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sensor measurements are used to evaluate far field plans, then route planning can be performed in uncharted environments, but measurement precision deteriorates leading to uncertainty and potential hazards
Solution Approach 1:
The patent segments the planning space into near field and far field regions. The near field uses high-fidelity sensor measurements for precise evaluation, while the far field uses lower-fidelity measurements with a coarser evaluation grid. This segmentation allows the system to maintain high measurement precision in critical near-field areas while still enabling route planning in uncharted far-field environments through acceptable lower-precision measurements.
2Measurement precision
If high fidelity measurements are used throughout the entire space, then measurement precision is improved, but use of energy increases and productivity decreases
Solution Approach 1:
The patent applies local quality by using high-fidelity sensor measurements and fine-grained evaluation only in the near field region where precision is critical for safety. In the far field region, the system transitions to lower-fidelity measurements and coarser evaluation grids, reducing computational load and energy consumption while maintaining sufficient planning capability. This localized approach optimizes the balance between measurement precision and route planning efficiency.
3Reliability
If comprehensive evaluation of all candidate plans is performed, then reliability is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary evaluation by first assessing far field plans using lower-fidelity measurements and coarser criteria to identify viable candidates. Only the most promising near field plans undergo comprehensive high-fidelity evaluation. This two-stage preliminary action filters out obviously poor plans early, reducing the time required for comprehensive evaluation while maintaining reliability through thorough assessment of final candidates.
Solution Approach 2:
The patent applies partial action by performing complete high-fidelity evaluation only for near field plans that pass initial far field screening. Far field plans receive partial evaluation using reduced-fidelity measurements and coarser grids, which is sufficient to eliminate poor candidates but unnecessary for final selection. This partial evaluation approach reduces overall computation time while maintaining plan reliability.
Data Source
AI summary
A plan through a space having a near field and a far field is determined. Using a sensor device, measurements of the far field are obtained and stored in an electronic memory. A processor uses the measurements to determine the viability of each far field plan among a plurality of candidate far field plans. The processor also determines a flexibility score for each of the candidate far field plans and selects a composite plan comprising the viable far field plan having a highest flexibility score among the viable candidate far field plans.