Dual GridTile Sensor Data Integration for Vehicle Assistance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driver assistance systems face significant discretization errors when integrating local sensor data into a global environment model, leading to inaccurate obstacle representation and increased complexity, particularly due to differences in cell orientation and size between local and global networks.

Innovation Solution

The system uses a dual GridTile structure where sensor data is stored in second virtual cells that match the dimensions and orientation of first virtual cells, allowing direct integration without additional discretization errors, using run-length coding for efficient data transfer and compression.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If local sensor networks are integrated into global environment models using traditional occupancy grids, then sensor data can be combined, but discretization errors increase and measurement precision deteriorates

Engineering Contradiction:
Improveintegration capabilityVSAvoidobstacle representation accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The environment is divided into discrete cells (GridTiles) that can be independently managed and integrated. Each sensor creates its own local GridTile representation, which is then segmented and merged into the global environment model, allowing precise obstacle representation while maintaining integration capability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical dimension to the traditional 2D occupancy grid by creating multiple levels of GridTiles (local sensor level and global environment level). This dimensional extension allows seamless integration of sensor data at different resolutions and orientations without losing precision

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If different sensor orientations are integrated into a global network, then comprehensive environment coverage is achieved, but transformation complexity and device complexity increase

Engineering Contradiction:
Improvesensor integration flexibilityVSAvoidtransformation rule complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The GridTile structure serves multiple functions simultaneously: it represents occupancy information, stores orientation data, and provides a universal interface for integrating sensors with different orientations. This universal structure eliminates the need for complex transformation rules while maintaining flexible sensor integration

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the parameters stored within each cell from simple occupancy values to include orientation and position information. This parameter enrichment allows the system to handle sensors with different orientations directly without requiring complex coordinate transformations

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If cell dimensions are increased to cover larger areas, then fewer cells are needed, but discretization error increases and measurement precision worsens

Engineering Contradiction:
Improvenumber of cellsVSAvoidobstacle representation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The environment is segmented into fine-grained GridTiles that can be densely packed when needed. This segmentation allows the system to use small cell dimensions for high precision obstacle representation while still managing computational complexity through efficient data structures

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts the effective resolution by activating only the GridTiles that are relevant to the current sensor field of view and vehicle position. This dynamic approach maintains high measurement precision in active areas while reducing overall computational complexity

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3008660B1Assistance device and method for assisting a driver of a vehicle
Publication Date: 2021.07.07 ROBERT BOSCH GMBH
  • EP3008660B1 patent drawingFigure 1~2
  • EP3008660B1 patent drawingFigure 3~4
  • EP3008660B1 patent drawingFigure 5~8

AI summary

The invention relates to an assistance device, in particular for a vehicle (1), comprising at least one sensor (2) for scanning a surrounding area (3) of the vehicle (1) and comprising at least one control device (4) which can exchange data with the sensor (2) in an alternating manner. The control device (2) has a first memory device which stores the scanned surroundings in at least one first virtual cell (5), in particular in at least one first GridTile, and the first virtual cell (5) represents a sub-region of the surrounding area (3) of the vehicle (1). The sensor (2) has a second memory device which stores data obtained by the sensor (2) in at least one second virtual cell (6), in particular a second GridTile, the second cell (6) represents at least one sub-region of a surrounding area (3) of the sensor (2) and at least partly represents the same region of the surroundings (3) as the first cell (5), and the control device (4) is designed to integrate the contents of the second cell (6) into the first cell (5).