Park Assist Route Training for Unmapped Valet Parking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced driver-assistance systems (ADAS) fail to provide valet parking assistance in areas without reliable digital maps or when existing maps are outdated, leaving users without the ability to use these features in desired locations.
Innovation Solution
A method involving a training mode where users manually drive through the parking area to create a trained digital map, using vehicle sensor data to define the parking route, which can then be used for automated drop-off and summon maneuvers, and optionally shared with others.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If ADAS uses existing digital map data for valet parking assistance, then the system can provide automated steering and maneuvering in mapped areas, but the system fails to provide functionality in areas without reliable or up-to-date digital maps
Solution Approach 1:
The system performs preliminary mapping actions by having the user manually drive through the parking area to capture sensor data and create a trained digital map before automated valet parking is needed. This preliminary action ensures the area is mapped with current and accurate information, allowing the system to provide reliable functionality in previously unmapped or outdated areas.
2Measurement precision
If the system requires pre-existing digital maps for valet parking, then map accuracy can be maintained, but users cannot use the feature in desired locations where maps are unavailable or outdated
Solution Approach 1:
The system enables users to create their own trained digital maps by manually driving through parking areas and capturing sensor data. This self-service approach allows users to generate accurate local maps on-demand for specific parking locations, expanding geographic coverage while maintaining map accuracy through direct user observation and sensor measurement.
3Reliability
If the system waits for crowd-sourced map data from multiple vehicles, then map coverage can be improved over time, but users must wait for critical mass of data collection before feature availability
Solution Approach 1:
Instead of waiting for crowd-sourced data accumulation, the system performs preliminary mapping immediately when needed by having the user drive through the area once to capture necessary sensor data. This eliminates the waiting time for critical mass data collection while still achieving reliable map coverage for the specific location.
4Extent of automation
If the system uses automated vehicle controls with sensors for drop-off and summon maneuvers, then driving automation can be achieved, but the system cannot operate in areas without proper digital map infrastructure
Solution Approach 1:
The system performs preliminary environmental scanning and map creation by having the user manually drive through the parking area and capture sensor data before automated operations are attempted. This preliminary action creates the necessary digital infrastructure (trained map) that enables high-level autonomous driving capabilities in previously unsupported environments.
Data Source
Figure 1
AI summary
Disclosed is a method of providing driver assistance in a vehicle, the method comprising the following steps: S1 entering a training mode; S2 while the vehicle is operating in training mode, acquiring and at least temporarily storing vehicle sensor data related to the vehicle's position during training mode; S3 at least partially based on the acquired vehicle sensor data, generating and storing a trained digital map of a parking route; S4 acquiring and storing position information related to a drop-off location and to a pickup location, the drop-off location and the pickup location being part of the parking route; and S5 exiting the training mode.