Autonomous Mobile Radar Fusion for 360-Degree Obstacle Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in autonomous mobile devices is achieving comprehensive obstacle sensing while minimizing costs and radar deployment, as well as integrating data from radars of different manufacturers with incompatible software development kits (SDKs).
Innovation Solution
The solution involves strategically arranging 3D all-round looking radars on an autonomous mobile device to ensure full 360-degree coverage with minimal radars, and using an Enhanced Communication Abstraction Layer (ECAL) protocol and a fusion module to integrate data from radars with different SDKs, transforming and fusing point cloud data into a common coordinate system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple radars from different manufacturers are deployed to ensure comprehensive obstacle sensing coverage, then the sensing coverage and reliability are improved, but the overall cost and device complexity increase
Solution Approach 1:
The patent implements a universal data integration architecture that can process and fuse data from radars of different manufacturers through a common interface layer. The controller is designed to handle heterogeneous radar data formats and coordinate systems, enabling the system to work with multiple radar types without requiring separate processing pipelines for each manufacturer, thus reducing overall system complexity while maintaining comprehensive coverage
Solution Approach 2:
The patent introduces an intermediary data processing layer that acts as a mediator between diverse radar sources and the control system. This intermediate layer standardizes data formats, performs coordinate transformations, and fuses data from multiple radars into a unified representation, thereby simplifying the integration process and reducing the complexity of direct multi-manufacturer radar deployment
2Reliability
If more radars are deployed to achieve full 360-degree coverage, then the obstacle detection capability is improved, but the overall cost increases
Solution Approach 1:
The patent merges data from multiple radars into a unified point cloud representation through coordinate system transformations and data fusion algorithms. By combining the sensing capabilities of fewer radars strategically positioned around the autonomous mobile device, the system achieves comprehensive 360-degree coverage equivalent to or better than using more radars, thereby reducing the total number of radars required while maintaining detection capability
Solution Approach 2:
The patent transforms radar data from different coordinate systems into a unified three-dimensional point cloud space centered on the autonomous mobile device. This dimensional transformation allows data from fewer radars to be integrated and visualized in a common spatial reference frame, maximizing the effective coverage and detection capability of each individual radar while reducing the total quantity needed
3Reliability
If radars from different manufacturers are integrated to provide more data sources, then the sensing performance is improved, but the data integration complexity due to incompatible SDKs increases
Solution Approach 1:
The patent segments the data integration process into distinct functional modules: a data acquisition layer that interfaces with specific manufacturer SDKs, a data standardization layer that transforms proprietary formats into a common structure, and a fusion layer that processes the standardized data. This segmentation isolates manufacturer-specific complexity to separate modules, allowing the core integration logic to remain simple while still supporting multiple radar sources with different SDKs
Solution Approach 2:
The patent changes the parameter representation of radar data by transforming coordinates, timestamps, and data formats into a standardized form. By applying parameter transformations that convert manufacturer-specific data representations into a universal format, the system can integrate data from different manufacturers without requiring complex custom integration logic for each source, thereby reducing integration complexity while maintaining sensing performance
Data Source
AI summary
The present disclosure relates to an autonomous mobile device, a control method applied to the autonomous mobile device, and a controller. In one aspect, the present disclosure provides the autonomous mobile device, which includes: a first radar; and a controller, configured to process first radar data from the first radar to operate the autonomous mobile device.