Lane Traffic Efficiency Evaluation for Stable Lane-Change Planning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing vehicle lane change intention determination methods based on machine learning algorithms are unstable due to the need for large amounts of historical data, leading to safety risks and instability in lane change planning.

Innovation Solution

A determination method and device for vehicle traffic efficiency that calculates lane traffic efficiency by assessing obstacle vehicles, including same-lane and different-lane leading vehicles, using velocity parameters and block cost coefficients to select a lane for lane changing, thereby improving the accuracy and stability of lane change planning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If machine learning algorithm is used for lane change intention determination, then prediction capability is improved, but stability deteriorates due to large amount of historical data required for training

Engineering Contradiction:

Improveprediction capabilityVSAvoidstability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the lane change intention determination into two independent parts: a prediction module that identifies potential lane change intentions, and a verification module that uses traffic efficiency parameters (calculated from velocity differences and block cost parameters) to verify and stabilize the determination. This segmentation allows the system to benefit from ML prediction while maintaining stability through traditional parameter-based verification.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces traffic efficiency parameters as an intermediary between the ML prediction and the final lane change execution. These parameters (velocity difference, block cost) act as a mediator that filters and stabilizes the ML output, preventing unstable or unsafe lane change intentions from being executed.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If machine learning algorithm is used for lane change intention determination, then prediction capability is improved, but safety risk increases due to instability

Engineering Contradiction:

Improveprediction capabilityVSAvoidsafety risk

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary anti-action by pre-calculating traffic efficiency parameters and block cost parameters before executing lane changes. The system proactively identifies and blocks unsafe lane change intentions by verifying that the target lane has sufficient traffic efficiency, thereby preventing safety risks before they occur.

Inventive Principle:

Principle #9Preliminary anti-action

Solution Approach 2:

The patent implements a feedback mechanism where the traffic efficiency parameters and block cost parameters continuously monitor and evaluate the safety of lane change intentions. The system uses this feedback to adjust and stabilize lane change decisions, ensuring that only safe intentions are executed.

Inventive Principle:

Principle #23Feedback

3Reliability

If traditional lane change determination is used, then stability is maintained, but recognition accuracy deteriorates

Engineering Contradiction:

ImprovestabilityVSAvoidrecognition accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent merges the advantages of both ML-based prediction and traditional parameter-based determination. The system combines ML prediction capability with traffic efficiency parameter calculation, creating a hybrid approach that achieves both high recognition accuracy and stability.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a composite determination system that integrates ML prediction algorithms with traditional traffic efficiency parameters. This composite approach combines the adaptability of ML with the stability of parameter-based methods, achieving superior overall performance.

Inventive Principle:

Principle #40Composite materials

Data Source

PatentUS20240290208A1Determination method and determination device for vehicle traffic efficiency, and vehicle

Publication Date: 2024.08.29 MOMENTA (SUZHOU) TECHNOLOGY CO LTD

AI summary

Provided are a determination method and a determination device for vehicle traffic efficiency, and a vehicle. The method includes: determining obstacle vehicles in lanes corresponding to a traveling vehicle, the obstacle vehicles including a same-lane leading vehicle and a different-lane leading vehicle of the traveling vehicle; determining a vehicle block cost parameter based on a first velocity of the traveling vehicle and a second velocity of the same-lane leading vehicle, and obtaining an obstacle block cost parameter of the different-lane leading vehicle; and determining lane traffic efficiency based on the vehicle block cost parameter, the obstacle block cost parameter, and a block weight coefficient, so as to select, based on the lane traffic efficiency, a lane for vehicle lane changing, the lane traffic efficiency representing an expected smooth traffic condition when the traveling vehicle enters a lane.