Sensor Fusion for Automated Driving Headway Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current automated driving systems face challenges in accurately controlling vehicle headway and predicting cut-in/cut-out scenarios, especially in complex driving conditions, which can lead to reduced reaction time and increased collision risks due to limitations in sensor accuracy and reliance on visible markers.

Innovation Solution

The implementation of a sensor-fusion-based system that utilizes behavioral prediction algorithms and multi-sensor data to assess collision threats and adjust vehicle speed or trajectory, enabling proactive preventive actions such as braking or speed reduction, and predicting vehicle cut-in/cut-out events to enhance safety and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If standard cruise control systems are used, then vehicle speed is maintained, but headway control accuracy deteriorates in complex driving conditions

Engineering Contradiction:

Improveheadway control accuracyVSAvoidperformance in complex driving conditions

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent combines multiple sensors (cameras, radar, LIDAR) to form a sensor fusion system that integrates data from different sources. This merging of sensing capabilities enables accurate headway control by compensating for individual sensor limitations and providing redundant information for more reliable detection and measurement in complex driving conditions.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces behavioral prediction algorithms as an intermediary layer between sensor detection and control execution. These algorithms process raw sensor data to predict future states of tracked objects, providing enhanced information that improves headway control accuracy while adapting to complex driving scenarios beyond simple marker detection.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If reliance is placed on visible markers for detection, then object detection is simplified, but reaction time decreases in cut-in/cut-out scenarios

Engineering Contradiction:

Improvedetection system complexityVSAvoidreaction time

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The patent implements behavioral prediction algorithms that perform preliminary analysis of tracked object trajectories and predict future cut-in/cut-out events before they occur. By anticipating potential hazards in advance rather than reacting to visible markers after they appear, the system gains valuable reaction time while maintaining manageable system complexity through algorithmic processing.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If sensor fusion is implemented, then collision avoidance capability is improved, but system complexity increases

Engineering Contradiction:

Improvecollision avoidance capabilityVSAvoidsensor fusion system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the sensor fusion system into distinct functional modules: data acquisition from multiple sensors, data processing and fusion algorithms, behavioral prediction subsystem, and control execution interface. This segmentation manages system complexity by organizing complex functions into manageable, independent modules that can be developed and maintained separately while working together to improve collision avoidance capability.

Inventive Principle:

Principle #1Segmentation

4Loss of time

If behavioral prediction algorithms are used, then time for evasive maneuvers is increased, but computational requirements increase

Engineering Contradiction:

Improvetime for evasive maneuversVSAvoidcomputational energy consumption

Core Design Contradiction:

Loss of timeVSUse of energy by moving object

Solution Approach 1:

The patent applies behavioral prediction algorithms selectively to tracked objects that present potential hazards, rather than performing exhaustive analysis on all detected objects. This partial action approach increases time for evasive maneuvers by focusing computational resources on critical predictions while reducing overall computational energy consumption by avoiding unnecessary processing of low-risk targets.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10864910B2Automated driving systems and control logic using sensor fusion for intelligent vehicle control

Publication Date: 2020.12.15 GM GLOBAL TECHNOLOGY OPERATIONS LLC

AI summary

Presented are automated driving systems for intelligent vehicle control, methods for making/using such systems, and motor vehicles equipped with such automated driving systems. A method for executing an automated driving operation includes: determining path plan data for a subject motor vehicle, including current vehicle location and predicted route data; receiving, from a network of sensing devices, sensor data indicative of current object position and object dynamics of a target object; applying sensor fusion techniques to the received sensor data to determine a threat confidence value that is predictive of target object intrusion with respect to the vehicle's location and predicted route; determining if this threat confidence value is greater than a calibrated threshold value; and, responsive to the threat confidence value being greater than the calibrated threshold value, transmitting one or more command signals to one or more vehicle systems (e.g., powertrain, steering and/or brake system) to take preventive action.