Maplets for Real-Time Digital Map Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital maps for autonomous vehicles and road networks are not updated in real-time, leading to inaccuracies and inefficiencies in navigation due to the lack of timely detection and communication of road changes and conditions.
Innovation Solution
A system that crowdsources data from sensors on vehicles to generate and transmit 'maplets' – standardized data packets containing real-time road information, using a predetermined data model and format to efficiently update digital maps, ensuring accurate and up-to-date navigation data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital maps are updated in real-time using crowdsourced sensor data from vehicles, then map accuracy and currency are improved, but system complexity and data processing requirements increase
Solution Approach 1:
Vehicles autonomously detect road features using their own sensors and generate maplet data without requiring centralized control or manual intervention. The system self-organizes by having each vehicle independently contribute to map updates as it traverses the road network, eliminating the need for complex centralized data collection infrastructure.
Solution Approach 2:
Instead of transmitting raw sensor data, the system creates standardized maplet representations that are simplified copies of the actual road environment. These maplets contain essential geometric and topological information in a standardized format, reducing data complexity while preserving accuracy needed for navigation.
2Measurement precision
If detailed sensor data is transmitted from vehicles to update maps, then map detail and accuracy improve, but network bandwidth consumption increases
Solution Approach 1:
The system extracts only the essential geometric and topological features needed for navigation from complete sensor data. Maplets contain selectively extracted information about road boundaries, intersections, and key features rather than transmitting all sensor measurements, significantly reducing data volume while maintaining navigation-relevant accuracy.
Solution Approach 2:
Sensor data is transformed from raw measurements into standardized maplet parameters representing geometric features. This parameter transformation compresses detailed sensor information into concise representations that convey essential spatial relationships with minimal data requirements.
3Reliability
If map data is updated frequently to reflect road changes, then navigation reliability improves, but processing time and computational resources increase
Solution Approach 1:
The road network is divided into discrete segments, each represented by individual maplet data structures. This segmentation allows the system to process and update only specific road segments that have changed rather than reprocessing entire maps, reducing computational overhead while maintaining up-to-date information where needed.
Solution Approach 2:
The system implements periodic validation and updating of map data through scheduled maplet requests and responses. Rather than continuous processing, vehicles periodically contribute maplet data and the system periodically validates and integrates changes, reducing processing time while ensuring maps remain current through regular updates.
Data Source
AI summary
An apparatus is onboard a vehicle and in communication with sensors onboard the vehicle. The apparatus receives a maplet request identifying a request region; and, responsive to determining that the vehicle is within the request region, processes sensor data captured by sensors to generate a multi-sensor data stream corresponding to a road network segment. The apparatus identifies an observation corresponding to a traffic signal within the multi-sensor data stream; and generates a maplet based on the observation and the maplet request. Generating the maplet comprises using a predetermined data model and a predetermined data format corresponding to a traffic signal observation class to encode road data corresponding to the observation corresponding to the traffic signal. The apparatus provides the maplet such that a network apparatus receives the maplet. The network apparatus is configured to validate/update map data of a digital map representing the road network based on the maplet.