Sensor-Based Autonomous Route Mapping With Real-Time Route Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in accurately constructing and updating driving routes, as they often rely on high-definition maps that can be expensive to maintain and may not accurately reflect real-world driving conditions.
Innovation Solution
A method for constructing autonomous driving routes using sensor-perceived information from autonomous mobility devices, which involves tracking moving and target objects, generating predicted routes, and updating route information based on actual movements to ensure accuracy and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-definition maps are used for autonomous driving, then route information precision is improved, but construction and update cost increases
Solution Approach 1:
The patent creates virtual copies of real driving routes by tracking and recording actual vehicle trajectories. These virtual route copies are then integrated into map information, providing accurate route data without requiring expensive manual HD map construction and updates. The system copies real-world driving behavior into digital map representations.
Solution Approach 2:
The system enables map information to self-update by automatically tracking vehicle movements and generating route information from actual driving data. This eliminates the need for manual map construction and updates, reducing costs while maintaining precision. The map information serves itself through automated data collection and processing from fleet vehicles.
2Productivity
If virtual driving routes are created on maps, then route information is generated, but deviation from real driving routes occurs
Solution Approach 1:
The system continuously tracks actual vehicle movements and compares them with virtual route predictions. This feedback loop allows the system to identify deviations between virtual and real routes, then correct the virtual route information to match actual driving behavior. The feedback mechanism ensures ongoing alignment between map information and real-world routes.
Solution Approach 2:
The system performs preliminary tracking and recording of actual driving routes to establish accurate baseline data before generating virtual route information. By capturing real route data in advance through vehicle tracking, the system creates more accurate virtual representations that closely match actual driving paths, reducing deviation.
3Ease of operation
If autonomous moving objects refer to virtual driving routes, then driving plans are executed, but malfunction or accident risk increases
Solution Approach 1:
The system dynamically updates virtual driving routes based on actual vehicle behavior and changing road conditions. Rather than relying on static virtual routes, the system adapts route information in real-time to reflect actual driving patterns and environmental changes, improving safety while maintaining ease of operation.
Solution Approach 2:
The patent merges virtual driving routes with actual driving route data from tracking systems. This combination creates hybrid route information that retains the planning benefits of virtual routes while incorporating the accuracy and adaptability of real-world driving data, thereby reducing malfunction and accident risk.
Data Source
AI summary
Disclosed herein a method for constructing autonomous routes based on sensor perceived information and a device for assisting autonomous driving. The method includes; obtaining a travel route by tracking a moving object and at least one target object among neighboring dynamic objects around the moving object based on perception information obtained from the moving object; generating a predicted route on which a movement is estimated from the travel route, based on the travel route; generating route information including the travel route and the predicted route and incorporating the route information into map information; and updating the route information based on actual information, in case there is a predetermined range or more of deviation between the actual information on an actual movement of the target object in a region corresponding to the predicted route and the route information.