Parking Lot Geometry Generation Using Probe Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional navigation systems and autonomous vehicles lack the necessary geometry data to guide vehicles through parking lots effectively, as existing map data often does not include detailed information about parking lot road networks, making it difficult to navigate to and from points-of-interest within these areas.
Innovation Solution
The creation of parking lot geometry using probe data to generate a road network, identify primary parking lot road links based on trajectory count, probe density, and average probe speed, and establish routes from parking lot accessors to building affinity links using a cost-based routing algorithm, enabling autonomous vehicle control and navigation through parking lots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional map data is used for navigation, then general road navigation is supported, but parking lot navigation and guidance to points-of-interest within parking lots is not available
Solution Approach 1:
The system performs preliminary actions by collecting and storing probe data from multiple vehicles traversing the parking lot before navigation is needed. This probe data including trajectories, speeds, and positions is pre-processed to generate the parking lot road network geometry, so that when a vehicle needs navigation, the geometry is already available without requiring real-time data collection
Solution Approach 2:
The system uses probe data from multiple vehicles as an intermediary to indirectly map the parking lot geometry. Instead of directly measuring the parking lot structure, the system collects trajectory data from vehicles that naturally traverse the space, then processes this intermediary data to reconstruct the road network geometry
2Measurement precision
If detailed parking lot geometry is created using probe data, then navigation accuracy within parking lots is improved, but data processing complexity and computational requirements increase
Solution Approach 1:
The system employs self-service mechanisms where multiple vehicles collectively contribute their own probe data to build the parking lot geometry. Each vehicle's trajectory data serves to refine and validate the overall geometry reconstruction, with the system automatically filtering and processing the data without requiring external intervention or manual surveying
Solution Approach 2:
The system uses excessive probe data from many vehicles to ensure accurate geometry reconstruction. By collecting more data than strictly necessary from multiple sources, the system can filter out anomalies and errors, achieving high measurement precision through redundancy rather than through complex single-source measurement systems
3Productivity
If a cost-based routing algorithm is used to establish optimal routes, then navigation efficiency is improved, but computational time for route calculation increases
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing the parking lot road network geometry with all necessary spatial relationships before navigation queries are made. This pre-processing of the geometric data structure enables faster route calculation when navigation is actually needed, as the foundational data is already organized and ready for query processing
Data Source
AI summary
A method is provided for the creation of parking lot geometry that can be used to provide guidance to a vehicle through the parking lot. Methods may include: generating a road network of a parking lot based on a plurality of probe data points; map matching the plurality of probe data points to road links of the parking lot; identifying parking lot accessor road links; identifying a building affinity link; identifying primary parking lot road links, where primary parking lot road links include at least one of a trajectory count, probe count, or average or median probe speed that is higher than a respective trajectory count, probe count, or average or median speed of non-primary parking lot road links; establishing a route from a parking lot accessor road link to the building affinity link; and providing for guidance of a vehicle along the route.