Autonomous Vehicle Lane Switching for Map Data Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Maintaining up-to-date map data for autonomous and semi-autonomous vehicles is challenging due to constantly changing road conditions, requiring frequent and costly data collection by dedicated vehicles.
Innovation Solution
Unoccupied autonomous vehicles select lanes for data collection based on criteria such as data staleness, construction, and defects, calculating the probability of finding a passenger to determine whether to switch from the curbside lane to a more data-worthy lane, thereby reducing the need for dedicated data collection vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If dedicated vehicles are dispatched to collect map data when it becomes outdated, then map data freshness is improved, but operational costs increase significantly
Solution Approach 1:
The patent applies multi-functionality by enabling autonomous vehicles to perform dual roles: (1) providing passenger transportation service and (2) collecting map data. Instead of using dedicated vehicles solely for data collection, the system utilizes existing autonomous vehicles that already traverse the road network, thereby eliminating the need for separate data collection fleets and reducing operational costs while maintaining map data freshness
Solution Approach 2:
The system implements self-service by having autonomous vehicles collect map data during their normal operational routes without requiring external dispatch or additional resources. The vehicles autonomously determine when and where to collect data based on their current location and the freshness requirements of map data, effectively serving their own data collection needs while providing transportation service
2Productivity
If autonomous vehicles travel in the curbside lane to maximize passenger pickup opportunities, then ride request probability is improved, but map data collection capability for other lanes deteriorates
Solution Approach 1:
The patent applies dynamics by making the lane selection of autonomous vehicles adaptive and variable rather than static. The system dynamically adjusts which lane a vehicle should be in based on real-time factors including the freshness of map data for different lanes, the probability of receiving ride requests in each lane, and the specific needs of the transportation service. This allows the system to optimize both passenger pickup efficiency and data collection coverage according to current conditions
Solution Approach 2:
The system implements preliminary action by proactively switching vehicles to non-curbside lanes when map data for those lanes is determined to be outdated or stale. Rather than waiting for data to become completely obsolete, the system anticipates data freshness requirements and adjusts vehicle positioning in advance, ensuring that critical map data is collected before it becomes unusable while minimizing impact on passenger pickup
Data Source
AI summary
In one embodiment, a method for collecting lane data is disclosed. The method includes collecting lane data for a first lane of a plurality of lanes of a road from one or more sensors of a vehicle traveling in the first lane. The method includes receiving an identifier of a lane-of-interest from the plurality of lanes. The method includes determining a probability that the vehicle will receive a ride request while traveling in the first lane. The method includes determining that the probability satisfies a threshold. The method includes, in response to determining that the probability satisfies the threshold, causing the vehicle to travel in the lane-of-interest. The method includes collecting lane data for the lane-of-interest from the one or more sensors.