Connected Vehicle Lane Selection Under Lane Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lane selection for vehicles is challenging due to limited vehicle view and sensor capabilities, and downstream events can affect performance, especially when high-definition maps and high-precision GPS are not available, which are costly and require frequent updates.
Innovation Solution
A decentralized lane selection technique using connected vehicle data that generates a dynamic lane-level forward graph, considers uncertainty, and computes weights based on metrics like congestion to select optimal lanes, utilizing a server to instruct vehicles on lane changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-definition map and high-precision GPS sensor are used to localize the vehicle and identify a lane, then lane selection accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent introduces a server as an intermediary that receives sensor data from vehicles, performs centralized lane identification and localization computations, and returns lane selection recommendations. This mediator approach allows vehicles to use simpler onboard sensors while achieving accurate lane identification through the server's processing capabilities, thus resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The system creates a virtual representation of the road environment and lane structure on the server, copying the essential geometric and topological information needed for lane identification. This virtual model allows the server to perform accurate lane localization without requiring each vehicle to possess expensive high-definition maps or precision GPS hardware.
2Measurement precision
If high-definition map and high-precision GPS sensor are used to localize the vehicle and identify a lane, then lane selection accuracy is improved, but cost increases
Solution Approach 1:
The server acts as a cost-effective intermediary that centralizes the computational burden and data processing requirements. Instead of equipping each vehicle with expensive high-definition maps and precision GPS sensors, the server performs these functions centrally using received sensor data, significantly reducing the cost per vehicle while maintaining lane identification accuracy.
Solution Approach 2:
The system enables vehicles to perform lane identification and localization using their own standard sensors by leveraging the server's processing capabilities. Each vehicle serves itself by sending its sensor data to the server, which returns lane identification results, eliminating the need for each vehicle to independently possess expensive specialized hardware.
3Device complexity
If vehicle systems utilize standard sensors without high-definition maps, then device complexity and cost are reduced, but lane selection accuracy deteriorates
Solution Approach 1:
The server serves as a compensatory intermediary that makes up for the limitations of standard onboard sensors. By receiving data from these simpler sensors and processing it against a virtual road model, the server recovers the lane identification accuracy that would otherwise require expensive specialized hardware, thus resolving the contradiction between device complexity and measurement precision.
4Device complexity
If centralized server processing is used for lane selection, then computational complexity in vehicles is reduced, but communication requirements increase
Solution Approach 1:
The patent extracts the computationally intensive tasks of lane identification, localization, and selection from the vehicle systems and relocates them to the centralized server. This extraction reduces vehicle computational complexity while the communication overhead is managed by transmitting only essential sensor data and receiving concise lane recommendation outputs, balancing the information exchange requirements.
Data Source
AI summary
A method may include generating a dynamic lane-level forward graph including a plurality of nodes for a road. The method may include computing weights for each action of a vehicle traveling from one node to a next node. The method may include determining values of different actions for the vehicle based on the dynamic lane-level forward graph starting from a node of the vehicle to a node of destination and the weights for each action of the vehicle. The method may include selecting an action among the different actions based on a comparison of the values of the different actions. The method may include instructing the vehicle to execute the selected action for the vehicle.