Adaptive Lane-Cell Modeling for Vehicle Tactical Environment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing environmental modeling systems for autonomous vehicles lack a tactical representation that adapts to the vehicle's traffic context and provide a detailed description of cell states, particularly for medium-term decision-making, as they often use fixed-dimension grids with limited states and do not account for vehicle interactions.
Innovation Solution
A method and device for modeling the tactical environment of an autonomous vehicle that constructs a geometric model of traffic lanes, estimates cell occupancy states, and defines additional states like neutralized and secure, using perception data and confidence indices to optimize discretization, ensuring accurate and reliable tactical decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed-dimension grid is used to model the environment, then the modeling process is simple, but the model cannot adapt to the vehicle's traffic context and provides insufficient detail for tactical decision-making
Solution Approach 1:
The patent applies dynamics by making the grid model adaptive rather than fixed. The system dynamically adjusts the grid dimensions and cell states based on the vehicle's traffic context, perception data, and tactical decision-making requirements. This allows the model to transform from a static structure to a dynamic one that evolves with changing traffic situations.
Solution Approach 2:
The patent implements parameter changes by modifying the grid model's parameters (dimensions, cell states, resolution) based on traffic context. The system changes parameters such as grid cell size, number of states per cell, and area of interest based on perception data and tactical requirements, enabling adaptation without complete model redesign.
2Loss of information
If only three states (free, occupied, undefined) are used in grid cells, then the model is simple, but it provides insufficient information for accurate tactical decision-making
Solution Approach 1:
The patent applies local quality by assigning different states to different cells based on their specific characteristics and relevance to tactical decision-making. Rather than using a uniform state system across all cells, the model assigns appropriate states (free, occupied, neutralized, secured, indeterminate) to each cell based on local traffic context, perception data, and tactical requirements.
Solution Approach 2:
The patent implements segmentation by dividing the environment into multiple grid cells, each with its own state. This segmentation allows the system to track and represent different regions of the environment with appropriate detail and state information, enabling comprehensive tactical awareness through distributed cell state representation.
3Measurement precision
If the model covers a large area for strategic decisions, then long-term planning is supported, but detail is lost for medium-term tactical decisions
Solution Approach 1:
The patent applies dynamics by making the grid model's area and resolution adaptive. The system dynamically adjusts the modeled area and cell resolution based on the decision-making horizon (tactical vs. strategic) and traffic context. This allows the model to transform from a fixed-area structure to a dynamic one that expands or contracts based on requirements.
Solution Approach 2:
The patent implements local quality by applying different levels of detail to different regions of the environment. Cells closer to the vehicle or in areas of tactical interest receive higher resolution and more detailed state information, while distant regions use coarser resolution, optimizing the balance between detail and coverage.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A method for modelling a tactical environment of a first motor vehicle, the first motor vehicle comprising: - a set of perception means, - a decision module, and - a module for geometric modeling of a set of traffic lanes (ENS_VOIES) to be taken into account by the decision module (3), the set of traffic lanes (ENS_VOIES) comprising first-order lanes (V11) interacting directly with a lane where the first motor vehicle (100) is travelling, and second-order lanes (V21, V22) interacting with the first-order lanes (V11), said modeling method comprising: - a first step (E1) of constructing, using the geometric modeling module (4), a geometric model (LGM) of a set of traffic lanes (ENS_VOIES) of the first order (V11) and of the second order (V21, V22), and of breaking the traffic lanes down into a set of cells (ENS_C) of identical length.