Robot Localization Maps Using BIM Semantics on Dynamic Sites
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots deployed in dynamically changing environments, such as construction sites, face challenges in maintaining accurate localization and navigation due to the presence of nonpermanent objects and changes in the environment's structure.
Innovation Solution
The use of a semantic model, specifically a Building Information Model (BIM), that includes semantic information about permanent objects within the environment, allows the robot to generate a localization map that distinguishes between permanent and nonpermanent features, ensuring reliable localization and navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot uses traditional localization methods in dynamic environments, then it can operate without semantic models, but the localization accuracy deteriorates due to nonpermanent objects and environmental changes
Solution Approach 1:
The patent introduces a semantic model as an intermediary layer between the robot's sensor data and the localization algorithm. This semantic model, which contains information about permanent objects in the environment, mediates the localization process by filtering and prioritizing features based on their permanence, thereby improving localization accuracy without requiring the robot to process all environmental features equally
Solution Approach 2:
The system performs preliminary action by pre-establishing the semantic model of the environment before the robot begins navigation. This model预先 identifies permanent objects and their locations, allowing the robot to focus its localization efforts on stable reference points rather than processing all environmental features in real-time, thus improving reliability without proportionally increasing operational complexity
2Reliability
If the robot builds a detailed localization map with all features, then the map provides comprehensive navigation information, but the map becomes unreliable as nonpermanent objects change the environment
Solution Approach 1:
The patent applies the extraction principle by separating permanent objects from nonpermanent objects in the environmental model. The semantic model extracts and isolates information about permanent structures, creating a distinct subset of the full environmental data that remains reliable over time. This allows the localization system to focus on permanent features while ignoring transient objects that would otherwise corrupt the map
Solution Approach 2:
The system applies local quality by assigning different reliability weights to different regions or features in the map based on their permanence. Permanent objects are marked with high reliability and serve as stable localization anchors, while areas with nonpermanent objects are marked with lower reliability. This localized quality assessment allows the map to maintain overall usefulness while accounting for local changes in the environment
3Measurement precision
If the robot manually updates the localization map frequently, then the map stays current with environmental changes, but the time and resources required increase significantly
Solution Approach 1:
The system implements self-service through automatic map updating mechanisms that use the semantic model to intelligently detect and process environmental changes. Rather than requiring frequent manual updates, the system automatically compares sensor data against the semantic model, identifies changes in permanent versus nonpermanent objects, and updates the localization map accordingly. This self-updating capability maintains map accuracy while eliminating the time and resources required for manual intervention
Data Source
AI summary
A method includes receiving, while a robot traverses a building environment, sensor data captured by one or more sensors of the robot. The method includes receiving a building information model (BIM) for the environment that includes semantic information identifying one or more permanent objects within the environment. The method includes generating a plurality of localization candidates for a localization map of the environment. Each localization candidate corresponds to a feature of the environment identified by the sensor data and represents a potential localization reference point. The localization map is configured to localize the robot within the environment when the robot moves throughout the environment. For each localization candidate, the method includes determining whether the respective feature corresponding to the respective localization candidate is a permanent object in the environment and generating the respective localization candidate as a localization reference point in the localization map for the robot.