Intersection Trajectory Coordination to Prevent AV Deadlock
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the number of ADAS-equipped vehicles increases, there is a need for systems and methods to prevent deadlock and gridlock situations without increasing complexity, while maintaining or improving vehicle operator comfort and ADAS functionality.
Innovation Solution
A system for autonomous vehicles that includes sensors, actuators, and control modules to predict trajectories, transmit messages, and coordinate actions to avoid deadlock at intersections using multi-directional conflicts identification and deadlock prevention coordination instructions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ADAS-equipped vehicles use traditional individual decision-making at intersections, then each vehicle can operate independently, but deadlock and gridlock situations occur when multiple vehicles encounter each other
Solution Approach 1:
The patent introduces a cloud-based server as an intermediary that receives trajectory predictions from multiple vehicles and generates coordinated control instructions. This mediator resolves conflicts between vehicles at intersections by computing optimal waiting relations and right-of-way assignments, preventing deadlock without requiring complex peer-to-peer negotiation protocols in each vehicle.
Solution Approach 2:
The system performs preliminary trajectory prediction and conflict identification before vehicles reach the intersection. By predicting potential deadlocks in advance and pre-coordinating waiting relations through the cloud server, the system prevents deadlock situations before they occur, rather than reacting after vehicles are already stuck.
2Reliability
If multiple ADAS-equipped vehicles coordinate their actions through cloud computing, then deadlock is prevented, but communication and processing time increases
Solution Approach 1:
The system performs trajectory prediction and conflict identification in advance before vehicles reach the intersection. By predicting potential deadlocks early and pre-coordinating through the cloud server, the system minimizes actual waiting time at the intersection while ensuring deadlock prevention.
Solution Approach 2:
Each vehicle independently predicts its own trajectory and transmits this information to the cloud server. The server then autonomously computes the coordination scheme and sends instructions back to vehicles, eliminating the need for lengthy multi-vehicle negotiation protocols and reducing overall coordination time.
3Reliability
If ADAS systems implement comprehensive trajectory prediction and coordination, then vehicle safety and deadlock prevention improve, but computational requirements and processing complexity increase
Solution Approach 1:
The system divides the computational workload into two segments: individual vehicles perform relatively simple trajectory prediction based on their own sensors and state, while the cloud server handles the complex multi-vehicle conflict resolution and coordination optimization. This segmentation prevents any single vehicle from needing excessive computational resources.
Solution Approach 2:
The cloud-based server acts as an intermediary that centralizes the complex computational tasks of analyzing multiple vehicle trajectories, identifying conflicts, and computing optimal coordination strategies. This moves the computational burden from individual vehicles to a centralized system with greater processing capacity.
4Adaptability or versatility
If autonomous vehicles use advanced coordination protocols, then ADAS functionality and redundancy are improved, but standardization requirements and implementation difficulty increase
Solution Approach 1:
The cloud-based coordination server provides a universal platform that can handle multiple vehicle types, different intersection configurations, and various traffic scenarios through a single standardized interface. Vehicles of different manufacturers and models can all interact with the same coordination system, reducing implementation complexity.
Solution Approach 2:
Each vehicle uses its existing sensors and onboard systems to independently generate trajectory predictions and receive coordination instructions. This self-service approach allows vehicles to implement the coordination protocol using their current hardware capabilities without requiring complex new components or extensive customization.
Data Source
AI summary
A system for deadlock precaution and prevention in autonomous vehicles (AVs) includes: AVs having actuators and sensors. The actuators alter a dynamic state of the AVs, and the sensors capture vehicle state and environmental information. The system receives vehicle state information and predicts a host AV trajectory, and trajectories of other vehicles approaching, entering, or within an intersection. Multi-directional conflicts identification per driving direction (MDCIPD) messages and vehicle state information are transmitted to a cloud computing device. Waiting relations for deadlock prevention are based on the MDCIPD messages and vehicle state information, and a deadlock precaution notification (DPN) and/or a deadlock prevention coordination instruction (DPCI) are generated. The AVs perform advanced driver assistance system (ADAS) functions via the actuators in response to the DPN and/or DPCI, and cause the AVs to avoid deadlock at the intersection.