Industrial Vehicle Landmark Localization for Start-Up Pose Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial vehicles face challenges in accurate localization at start-up within physical environments lacking unique natural features, which is crucial for safe and efficient navigation and task execution.
Innovation Solution
The use of unique landmarks, such as pre-positioned objects or markers, sensed by sensor arrays and processed by mobile computers to determine the vehicle's pose and navigate it accurately within the environment, enabling precise positioning and task completion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If industrial vehicles rely on natural features for localization, then they can determine position in environments with sufficient features, but they fail to localize accurately in environments with limited unique natural features
Solution Approach 1:
The system performs preliminary localization actions by first attempting to detect natural features, and only when those fail does it proceed to use artificial landmarks. This preliminary detection approach allows the vehicle to adapt to different environment types without compromising localization accuracy in either scenario.
Solution Approach 2:
Artificial landmarks serve as an intermediary solution when natural features are insufficient. The system introduces these man-made markers as a mediator between the vehicle's localization needs and the environment's limited natural features, enabling accurate positioning in previously unsuitable environments.
2Productivity
If warehouses add more forklifts and drivers to increase productivity, then task completion speed increases, but safety risks and operational complexity increase
Solution Approach 1:
The patent replaces manual forklift operation with autonomous vehicles equipped with sensor arrays and mobile computers. This substitution eliminates human drivers from the equation, removing the safety risks associated with human operation of heavy machinery while maintaining or improving productivity through automated task execution.
Solution Approach 2:
The autonomous forklifts perform self-localization and self-navigation using onboard sensors and processed landmark data. Each vehicle independently determines its own position and plans its path without human intervention, enabling the system to scale productivity by simply adding more autonomous units without proportionally increasing operational complexity.
3Measurement precision
If sensor arrays are mounted on moving parts like lift carriages, then localization data is collected from multiple positions, but the system complexity and cost increase
Solution Approach 1:
The sensor array system is designed to function effectively whether mounted on stationary or moving parts of the forklift. The same sensors can collect localization data from various positions during different operational states, making the system universally applicable without requiring separate sensor systems for different mounting locations.
Solution Approach 2:
The system dynamically adapts to sensor mounting locations whether on the body or moving parts. By processing landmark detection data in conjunction with real-time position information from encoders and other sensors, the mobile computer can accurately determine pose regardless of where sensors are physically located on the vehicle structure.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of operating an industrial vehicle comprising a lift carriage which comprises one or more lifting elements is provided. The industrial vehicle is coupled to a sensor array and a mobile computer which are communicatively coupled together. The sensor array comprises a plurality of sensor devices, at least one of which is mounted to the lift carriage. The method comprises using the mobile computer to: receive measurement data from the sensor array; determine the pose of the industrial vehicle in an environment based on this data; and navigate the industrial vehicle to a pallet or product load using the determined pose. The method further comprises using the mobile computer to position the industrial vehicle to pick up the pallet or product load; and sense a unique identifying feature comprised on the pallet or product load using the at least one sensor device. When the sensing device senses the unique identifying feature, pre-positioned object data associated with the pallet or product load is identified and used to develop a new pose of the industrial vehicle. The industrial vehicle is subsequently navigated through the environment using the new pose and data from the sensor array.