Pallet Face Tracking for Forklift Engagement Misalignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles in warehouses often inadvertently displace pallets during engagement or disengagement due to errors in positioning, leading to improper pallet handling and potential damage, despite tines being matched to pallet openings.
Innovation Solution
A control system uses sensor data to determine a baseline geometric representation of a pallet's face, tracking its pose and adjusting vehicle motion to avoid unplanned displacement by comparing updated geometric representations to the baseline, utilizing exclusion fields to detect and mitigate unintended repositioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If autonomous vehicles use standard positioning methods to engage pallets, then the engagement process is simple and fast, but positioning errors cause inadvertent pallet displacement and damage
Solution Approach 1:
The system performs preliminary actions by establishing a baseline geometric representation of the pallet's initial pose before engagement occurs. This baseline serves as a reference for detecting any unintended displacement during the engagement and disengagement processes, allowing the system to prevent damage before it happens.
Solution Approach 2:
The control system continuously monitors the pallet's pose during engagement and disengagement by comparing updated geometric representations against the baseline. When deviations exceeding a threshold are detected, the system provides feedback to modify vehicle motion, preventing inadvertent displacement and potential damage.
2Productivity
If the vehicle moves quickly to engage or disengage pallets, then productivity is improved, but positioning accuracy decreases leading to pallet displacement
Solution Approach 1:
The real-time feedback mechanism allows the system to maintain high productivity by enabling fast engagement and disengagement movements while continuously monitoring pallet position. The feedback loop detects deviations and triggers corrective actions only when necessary, allowing normal high-speed operations to proceed without interruption.
Solution Approach 2:
The system dynamically adjusts the engagement process by modifying vehicle motion in real-time based on detected pallet position deviations. This dynamic adjustment allows the system to maintain high speeds during normal operations while automatically slowing down or correcting course when positioning accuracy becomes compromised.
3Reliability
If geometric representations are continuously monitored during engagement, then pallet displacement is detected early, but computational load and processing time increase
Solution Approach 1:
The system applies local quality by focusing computational resources on monitoring only the critical geometric features of the pallet that are most likely to indicate displacement. Rather than analyzing the entire pallet structure, the system concentrates on key reference points and planes that provide sufficient information to detect unintended movement.
Solution Approach 2:
The system uses partial monitoring by establishing a baseline representation and only continuously comparing critical updated geometric features against this baseline. Full comprehensive monitoring is not performed at all times; instead, the system monitors selectively based on the engagement state, reducing computational load while maintaining sufficient detection reliability.
Data Source
Figure 1A
Figure 1B
Figure 2A~2B
AI summary
An example method may include receiving, from a sensor on a vehicle, an initial plurality of sensor data points representing a position of a face of a pallet. The vehicle may include tines configured to engage the pallet. A baseline geometric representation of the face of the pallet may be determined based on the initial plurality of sensor data, points. The vehicle may be caused to reposition the tines relative to the pallet. A subsequent plurality of sensor data points representing the position of the face of the pallet after repositioning the tines may be received from the sensor. An updated geometric representation of the face of the pallet may be determined based on the subsequent sensor data points. It may be determined that the updated geometric representation deviates from die baseline geometric representation by more than a threshold value and, in response, motion of the vehicle may be adjusted.