Autonomous Driving Model Switching for Event Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional autonomous driving systems struggle to rapidly adapt to unexpected events such as traffic accidents, road loss, or system errors, leading to potential accidents and safety risks for vehicle occupants.

Innovation Solution

An autonomous driving apparatus equipped with multiple pre-learned driving models that can be dynamically switched based on real-time sensory data, allowing the vehicle to adjust its driving mode in response to various events, such as accidents, weather conditions, or system abnormalities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a single autonomous driving model is used for ordinary driving situations, then the system is simple and easy to operate, but it cannot rapidly adapt to unexpected events such as accidents, road loss, or system errors

Engineering Contradiction:
Improveadaptability to eventsVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The autonomous driving system is segmented into multiple specialized models: a first autonomous driving model for normal driving situations and at least one second autonomous driving model for specific event situations (accidents, road loss, system errors). This segmentation allows each model to be optimized for its specific function while maintaining overall system adaptability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically switches between different autonomous driving models based on the detected situation. The processor determines whether an event has occurred and selects the appropriate model (first or second) accordingly, enabling the system to adapt its behavior in real-time without requiring complete reconfiguration.

Inventive Principle:
Principle #15Dynamics

2Reliability

If multiple autonomous driving models are prepared for different events, then the system can rapidly adapt to various situations, but the device complexity increases

Engineering Contradiction:
Improvesafety during autonomous drivingVSAvoidnumber of driving models
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Multiple autonomous driving models are prepared in advance for different event scenarios. The processor determines whether an event has occurred and selects the appropriate pre-prepared model, enabling rapid response without requiring real-time model generation or complex decision-making during critical events.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The autonomous driving apparatus is designed with multi-functionality by incorporating both a first autonomous driving model for normal operations and at least one second autonomous driving model for event-specific situations. This universal design allows a single system to handle diverse driving conditions effectively.

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

3Reliability

If the autonomous driving model is changed rapidly in response to events, then safety is improved, but the system requires more complex switching mechanisms

Engineering Contradiction:
Improveresponse to eventsVSAvoidmodel switching mechanism
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system employs dynamic model switching where the processor continuously monitors for events and transitions between the first autonomous driving model and at least one second autonomous driving model based on detected conditions. This dynamic approach enables rapid adaptation while maintaining a relatively simple switching mechanism driven by event detection.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback from event detection to trigger model switching. The processor determines whether an event has occurred based on sensor data and system state, then selects the appropriate autonomous driving model accordingly. This feedback-driven switching mechanism simplifies the control logic while ensuring appropriate response to events.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3676147B1Autonomous driving apparatus and method for autonomous driving of a vehicle
Publication Date: 2022.04.27 SAMSUNG ELECTRONICS CO LTD
  • EP3676147B1 patent drawingFigure 1~3
  • EP3676147B1 patent drawingFigure 4~5
  • EP3676147B1 patent drawingFigure 6~8a

AI summary

A photographing apparatus for preventing or reducing light leakage and an image sensor thereof are provided. The photographing apparatus includes an image sensor configured to include a plurality of pixels respectively having a Photo Diode and a Storage Diode for temporarily storing a charge accumulated in the Photo Diode and an image processor configured to perform an image processing operation by receiving the charge stored in the Storage Diode of each of the plurality of pixels. In addition, the image sensor has a structure where the Storage Diodes of the plurality of pixels are arrayed to be adjacent to each other. Accordingly, the photographing apparatus may prevent the light leakage from the adjacent pixel being flowed into a Storage Diode of each pixel.