Vehicle Interaction Control for Priority-Based Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous driving, vehicles often experience false deceleration and acceleration when trajectory conflicts occur, affecting safe travel and driving experience due to inadequate decision-making in resolving vehicle interactions.
Innovation Solution
A control method that classifies vehicles into active and passive parties based on passage priority, determines feasible strategies for both, calculates execution costs, and selects the strategy with the lowest cost to make preemption or yielding decisions, ensuring clear intention probing before executing maneuvers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vehicles perform false deceleration and false acceleration when trajectory conflicts occur, then collision avoidance is achieved, but driving experience and safety are degraded
Solution Approach 1:
The system performs preliminary classification of vehicles into active and passive parties based on passage priority before conflict resolution. This preliminary categorization enables the active party to maintain its trajectory while the passive party yields, avoiding false deceleration/acceleration maneuvers and improving driving experience without compromising collision avoidance
Solution Approach 2:
The system changes the decision-making parameter from reactive speed adjustments (false deceleration/acceleration) to proactive role-based strategy selection. By introducing execution cost calculations that consider comfort, time, and energy parameters, the system selects optimal strategies that maintain safety while enhancing driving experience
2Reliability
If vehicles make immediate preemption or yielding decisions when trajectory conflicts occur, then collision avoidance is achieved, but driving experience is degraded due to hard braking
Solution Approach 1:
The system introduces execution cost calculation as a cushioning mechanism before final decision execution. By evaluating comfort costs, time costs, and energy costs in advance, the system selects strategies that minimize hard braking and other harmful maneuvers while ensuring collision avoidance, thereby reducing negative impacts on passengers and surrounding environment
3Measurement precision
If vehicles continuously probe traveling intention of another object, then preemption or yielding decision accuracy is improved, but response time is increased
Solution Approach 1:
The system performs preliminary classification of vehicles into active and passive parties based on passage priority before detailed intention probing. This preliminary categorization narrows down the decision space and enables faster intention recognition, reducing response time while maintaining accuracy through subsequent execution cost evaluation
Solution Approach 2:
The system applies different probing depths to different scenarios. For clear active-passive relationships, minimal probing is needed. For ambiguous situations, the system intensifies intention probing through execution cost calculation of multiple strategies, achieving accurate intention recognition efficiently by allocating computational resources locally where needed
Data Source
AI summary
A method includes: when there is a possibility of collision between a target vehicle and a target object, classifying the target vehicle and the target object into an active party and a passive party, where the active party has a passage priority relative to the passive party; obtaining a first strategy set that is feasible for the active party; determining a second traveling strategy of the passive party corresponding to each first traveling strategy in the first strategy set based on the first strategy set, to obtain a second strategy set; analyzing execution costs of feasible strategies obtained based on the first strategy set and the second strategy set, to obtain a first cost set; and controlling the target vehicle according to a traveling strategy corresponding to a lowest execution cost in the first cost set.