Vehicle Behavior Planning Using Geometric Drivable Surface Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicle behavior planning methods lack precision in distinguishing drivable and non-drivable surfaces, leading to potential collisions and inefficient route planning, especially in complex traffic scenarios.

Innovation Solution

A method that receives environmental data and route information to ascertain geometric behavior options, defining drivable, non-drivable, and to-be-driven surfaces based on traffic scenarios, allowing vehicles to plan behavior exclusively within safe, drivable areas, using a combination of geometric and semantic information to ensure safe navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If current vehicle behavior planning methods are used, then the system is simple to operate, but the precision in distinguishing drivable and non-drivable surfaces deteriorates, leading to potential collisions

Engineering Contradiction:
Improveprecision in distinguishing drivable and non-drivable surfacesVSAvoidcomplexity of behavior planning system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The environment is segmented into distinct drivable and non-drivable surfaces through geometric behavior options. The behavior planning is divided into multiple independent geometric options, each defining specific drivable surfaces, allowing precise distinction without requiring a monolithic complex system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different geometric behavior options provide local quality variations in surface classification. Each behavior option defines specific drivable surfaces appropriate for particular traffic scenarios, enabling precise local discrimination of drivable areas without uniform system complexity throughout.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If geometric behavior options are ascertained based on traffic scenarios, then the precision of behavior planning is improved, but the computational complexity increases

Engineering Contradiction:
Improveprecision of behavior planningVSAvoidcomputational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

Geometric behavior options are ascertained in advance based on traffic scenarios before actual navigation decisions are made. This preliminary classification of drivable surfaces for different traffic situations enables precise behavior planning without requiring complex real-time computations during execution.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If vehicles plan behavior exclusively within drivable surfaces, then collision safety is improved, but the efficiency of route planning deteriorates due to restrictions

Engineering Contradiction:
Improvecollision safetyVSAvoidefficiency of route planning
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system dynamically selects from multiple geometric behavior options based on the current traffic scenario. Each option defines different drivable surfaces, allowing the vehicle to adapt its navigation constraints dynamically rather than following fixed restrictions, thereby maintaining both safety and route planning efficiency.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS20240208504A1Method for behavior planning of a vehicle
Publication Date: 2024.06.27 ROBERT BOSCH GMBH
  • US20240208504A1 patent drawing
  • US20240208504A1 patent drawing
  • US20240208504A1 patent drawing

AI summary

A method for behavior planning of a vehicle. The method includes: receiving data of an environment detection system and of a planned route of the vehicle; ascertaining a traffic scenario based on the data of the environment detection system and of the planned route; ascertaining at least one geometric behavior option based on the traffic scenario; ascertaining a behavior that can be performed by the vehicle, taking into account the at least one behavior option.