Adaptive Map Selection for Real-Time Autonomous Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous navigation systems face challenges in providing accurate and flexible map information suitable for diverse autonomous navigation devices, leading to inefficiencies and potential malfunctions or accidents due to discrepancies between actual and preset movement paths.
Innovation Solution
A method and device for providing an adaptive map that dynamically applies map information in real-time, selecting and adjusting map information based on the characteristics of the moving object and its environment, using AI-based algorithms to ensure flexibility and safety in various situations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HD map with high accuracy and large amount of information is used, then navigation accuracy is improved, but resource consumption increases and real-time update speed decreases
Solution Approach 1:
The map information is segmented into multiple types (geometric map, topological map, semantic map, occupancy map, cost map) based on different navigation needs. Each map type contains specific information for particular functions, allowing the system to select and process only relevant map segments rather than handling all HD map data, thus reducing resource consumption while maintaining navigation accuracy.
Solution Approach 2:
The system dynamically adjusts the amount and type of map information provided to autonomous vehicles based on their specific characteristics (sensing performance, processing performance, communication performance) and current environmental conditions. This dynamic adaptation enables real-time optimization of resource allocation, updating map information at appropriate speeds matching each vehicle's capabilities rather than using a fixed high-update-rate approach for all vehicles.
2Measurement precision
If map information amount is increased to prevent fragmentation, then navigation accuracy is improved, but processing speed and communication efficiency decrease
Solution Approach 1:
Different map information types are provided with different levels of detail and update frequencies based on local requirements. For example, geometric maps may be updated at different rates than semantic maps depending on the vehicle's processing capabilities and the specific navigation task, allowing high accuracy where needed while maintaining efficient processing overall.
3Adaptability or versatility
If diverse map data formats and communication protocols are supported for different autonomous driving systems, then system versatility is improved, but service integration difficulty increases
Solution Approach 1:
The server is designed with universal functionality to support multiple autonomous vehicle types (ground vehicles, aerial vehicles, marine vehicles) with different sensing, processing, and communication capabilities. The system provides a unified interface that adapts to various vehicle requirements through standardized map information types, enabling one system to serve multiple functions and vehicle platforms without requiring separate specialized systems for each vehicle type.
Data Source
AI summary
Disclosed herein a method and device for providing an adaptive map for diverse autonomous navigation platforms. The method includes: generating environment information around a moving object based on perception information acquired from the moving object equipped with a sensor; selecting at least one piece of map information among multiple pieces of map information based on characteristics information of a type of the moving object, the moving object, which includes at least one of sensing performance of the moving object, processing performance of the moving object, and communication performance of the moving object, and the environment information; adjusting the map information based on situation information representing a situation around the moving object recognized from the environment information; and transmitting the adjusted map information to the moving object.