Warehouse Robot Semantic Localization Without Pre-Arranged Signs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing warehouse localization methods rely heavily on pre-arranged signs, leading to poor localization flexibility and inefficiency.
Innovation Solution
A method for constructing a semantic map using visual semantic localization, where a warehousing robot detects and recognizes objects in its environment to determine its location based on relative position relationships and semantic information, eliminating the need for pre-arranged signs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If signs like reflective strips and two-dimensional codes are pre-arranged in the warehouse environment for localization, then localization accuracy can be improved, but localization flexibility deteriorates due to high dependence on these signs
Solution Approach 1:
The patent extracts the localization function from dependence on pre-arranged signs and transfers it to natural warehouse objects. Instead of requiring reflective strips or two-dimensional codes, the system detects and uses existing objects like shelves, boxes, and pallets as localization landmarks, thereby eliminating the harmful dependence on artificial signs while maintaining localization accuracy.
Solution Approach 2:
The patent makes the localization system universal by enabling it to work with various types of warehouse objects (shelves, boxes, pallets, containers) without requiring specific pre-arranged signs. The semantic map can store multiple types of objects with different features, allowing the robot to adapt to different warehouse environments and object configurations, thus improving both flexibility and adaptability.
2Reliability
If pre-arranged signs are used for localization, then localization can be achieved, but system complexity increases due to the need for sign installation and maintenance
Solution Approach 1:
The patent applies self-service by enabling the warehouse environment itself to provide localization features through its natural objects. The shelves, boxes, and pallets automatically serve as localization landmarks without requiring external installation or maintenance of special signs. The robot's image sensor and processing system autonomously detect and utilize these objects, eliminating the need for separate sign installation and maintenance processes.
Solution Approach 2:
The patent replaces the mechanical system of physical sign installation and maintenance with an optical and computational system. Instead of physically installing reflective strips or two-dimensional codes on warehouse objects, the system uses image sensors to capture visual information and computational algorithms to extract localization features from natural objects, thereby reducing mechanical complexity while maintaining localization reliability.
3Adaptability or versatility
If visual semantic localization with object detection is implemented, then localization flexibility improves by eliminating sign dependence, but measurement precision may deteriorate without artificial signs
Solution Approach 1:
The patent changes the parameters used for localization from artificial sign features (reflective strips, two-dimensional codes) to natural object features (shape, size, position, semantic information of shelves, boxes, pallets). By transforming the feature extraction process to utilize rich semantic information from deep learning object detection, the system maintains high localization accuracy while working with natural warehouse objects instead of artificial signs.
Solution Approach 2:
The patent combines multiple types of information (image data, depth information, semantic labels, position coordinates) to create a composite localization system. By fusing visual semantic information from object detection with spatial relationship data from the semantic map, the system achieves high localization accuracy using natural warehouse objects, compensating for the absence of artificial signs through multi-source information integration.
Data Source
AI summary
Disclosed are a method for locating a warehousing robot, a method for constructing a map, a robot and a storage medium. In a specific embodiment, a semantic map of a warehouse environment is constructed in advance, and the semantic map comprises a plurality of objects existing in the warehouse environment and semantic information of the objects. In the localization process, a warehousing robot uses its own image sensor to acquire an image or video data of a surrounding environment (11), identifies target objects in the image or video data and semantic information of the target objects (12) to obtain the relative position relationship between each target object and the warehousing robot (13), and then determines the location of the warehousing robot in the semantic map based on the relative position relationship and the semantic information of each target object (14). The method for constructing a map is based on visual semantic localization. Because the method directly detects specific targets, the detection speed is fast, semantic information is rich, and the method is not easily influenced by other interference factors. The method gets rid of the dependence on signs in the warehouse environment and has high localization flexibility.