Autonomous Vehicle Turn Priority Bidding at Occluded Intersections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion planning techniques for autonomous vehicles are computationally intensive and inefficient, particularly at intersections, due to reliance on sensor signals and traditional right-of-way rules, which can be affected by occlusions and are not optimized for autonomous vehicle environments.
Innovation Solution
An autonomous vehicle system that generates bids for turn priority based on trip characteristics, such as passenger importance and energy efficiency, using a networked computing system to determine a turn order that optimizes traffic flow and reduces computational burden, allowing vehicles to navigate intersections more efficiently without relying solely on traditional right-of-way rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motion planning techniques process a large amount of sensor signals to determine right of way, then the determination accuracy is improved, but the computational burden increases
Solution Approach 1:
The patent extracts and separates the right of way determination logic from conventional motion planning. Instead of processing all sensor signals through complex motion planning algorithms, the system extracts specific intersection-related sensor signals and processes them through a dedicated right of way determination module, reducing overall computational burden while maintaining determination accuracy
Solution Approach 2:
The patent segments the autonomous driving system into distinct functional modules: conventional motion planning for general navigation and a specialized right of way determination module for intersection scenarios. This segmentation allows each module to focus on specific tasks, improving efficiency without sacrificing accuracy
2Reliability
If conventional motion planning techniques rely on sensor signals, then the right of way determination is made, but occlusions in driving environments prevent perception of certain areas, negatively affecting determination
Solution Approach 1:
The patent introduces virtual sensor signals as an intermediary to compensate for physical sensor limitations. When occlusions prevent direct perception, the system generates virtual sensor data based on historical data, environmental context, and predictive models, allowing the right of way determination to proceed reliably despite physical sensing limitations
Solution Approach 2:
The patent creates virtual copies of sensor signals through simulation and historical data reconstruction. These virtual sensor signals replicate the information that would be obtained if occlusions were not present, enabling the determination system to make reliable decisions based on complete environmental information
3Ease of manufacture
If conventional motion planning techniques use traditional right of way rules, then the system is simple to implement, but it is not efficient or desirable in environments with many autonomous vehicles
Solution Approach 1:
The patent transitions from static, rule-based right of way determination to a dynamic, adaptive system. The determination module continuously evaluates multiple factors including vehicle trajectories, passenger importance, environmental context, and real-time sensor data, allowing the system to optimize intersection traversal efficiency dynamically while maintaining implementation feasibility
Solution Approach 2:
The patent changes the parameters used for right of way determination from simple rule-based criteria to a multi-parameter evaluation system. This includes considering passenger importance levels, vehicle speed, acceleration, environmental factors, and predictive trajectory data, enabling efficient intersection navigation in autonomous vehicle environments
Data Source
AI summary
Various technologies described herein pertain to generating a bid for turn priority at an intersection. An autonomous vehicle determines that the autonomous vehicle and a second autonomous vehicle are proximate to an intersection. The autonomous vehicle generates a first bid that is indicative of a first importance that the autonomous vehicle traverses the intersection. The first bid is based upon characteristics of a trip of a passenger riding in the autonomous vehicle. The autonomous vehicle transmits the first bid to a networked computing system, wherein the networked computing system determines a turn order based upon the first bid and a second bid generated by the second autonomous vehicle. The networked computing system transmits the turn order to the autonomous vehicle, wherein the autonomous vehicle operates based upon the turn order.