HD Map Synthetic Sensor Data for Autonomous Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maps for autonomous vehicles lack precision and accuracy, leading to challenges in navigation due to obscured sensor inputs and outdated data, which can result in safety issues and inefficiencies in updating road information.
Innovation Solution
The use of high-definition (HD) maps that allow for real-time data collection and updating through vehicle sensors, enabling precise location determination and safe navigation by generating synthetic sensor data and incorporating synthetic objects to simulate scenarios like lane closures, thereby improving navigation accuracy and data freshness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional maps are created by survey teams using specially outfitted survey cars, then map data can be collected, but the process is expensive and time-consuming, resulting in outdated map information that cannot keep up with frequent road changes
Solution Approach 1:
The patent creates synthetic copies of sensor data by rendering virtual sensor measurements from HD map data and synthetic objects. This copying approach allows generating unlimited training scenarios without deploying physical survey cars, resolving the contradiction between measurement precision and update frequency by enabling rapid map updates through virtual replication rather than expensive physical surveys
Solution Approach 2:
The system uses the autonomous vehicles' own sensor data and computing resources to continuously update and validate HD maps. Vehicles self-service by contributing their sensor measurements to map updates, eliminating the need for dedicated survey teams while maintaining high accuracy and frequent updates through distributed crowd-sourced data collection
2Ease of operation
If GPS systems are used for location determination, then basic navigation is provided, but accuracy is limited to approximately 3-5 meters with large error conditions resulting in over 100 m inaccuracies
Solution Approach 1:
The patent merges GPS data with HD map data and synthetic sensor data to create a multi-source localization system. By combining the broad coverage of GPS with the high-precision geometric constraints of HD maps and synthetic sensor simulations, the system achieves both ease of operation and high measurement precision, overcoming the limitations of GPS alone
Solution Approach 2:
The HD map serves as an intermediary between GPS and the vehicle's actual position. The synthetic sensor data acts as a mediator that translates GPS coordinates into precise vehicle poses by comparing virtual sensor readings from map positions with actual sensor data, achieving high location accuracy while maintaining operational simplicity
3Reliability
If vehicle sensors are used to detect road inputs, then real-time data is collected, but sensors may be obscured by corners, rolling hills, other vehicles, or may not observe certain inputs early enough to make safe driving decisions
Solution Approach 1:
The system performs preliminary actions by pre-rendering synthetic sensor data for future vehicle positions and potential scenarios. By pre-computing sensor readings for upcoming road sections and synthetic objects, the system prepares advance information that compensates for current sensor obscurations and reduces response time, maintaining reliability while overcoming time delays
Solution Approach 2:
The patent creates virtual copies of sensor data through rendering synthetic measurements from HD map positions. These synthetic sensor copies provide backup detection capabilities when physical sensors are obscured, maintaining reliable detection by substituting virtual sensor readings that are not subject to physical obstructions or timing delays
4Productivity
If conventional maps lack precision and accuracy, then basic navigation is possible, but navigation safety and efficiency are compromised due to obscured sensor inputs and outdated data
Solution Approach 1:
The patent creates synthetic copies of road features and sensor data to enhance conventional map information. By rendering virtual representations of road geometry, signs, and obstacles from HD map data, the system produces high-precision navigation information that improves both safety and efficiency without requiring expensive physical surveying for every update
Solution Approach 2:
The system changes the precision parameters of map data by generating HD-level detail from conventional map inputs. Through synthetic data rendering and multi-source fusion, the patent transforms low-precision conventional maps into high-precision navigation data, simultaneously improving navigation safety and efficiency through enhanced spatial accuracy and up-to-date information
Data Source
AI summary
According to an aspect of an embodiment, operations may comprise accessing high definition (HD) map data of a region, presenting, via a user interface, information describing the HD map data, receiving instructions, via the user interface, for modifying the HD map data by adding one or more synthetic objects to locations in the HD map data, modifying the HD map data based on the received instructions, and generating a synthetic track in the modified HD map data comprising, for each of one or more vehicle poses, generated synthetic sensor data based on the one or more synthetic objects in the modified HD map data.