Autonomous Driving Path Guidance for Undefined Objects and Intent

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data-driven autonomous driving methods struggle to recognize undefined objects and identify the intentions of pedestrians and surrounding vehicles, particularly in novel or unlearned driving situations, limiting their ability to respond effectively.

Innovation Solution

A knowledge-driven autonomous driving method that utilizes a generative neural network model to generate guide information based on encoded sensor data, incorporating past driving experience and accumulated knowledge to enhance recognition and decision-making in various driving scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data-driven artificial neural network models are used for autonomous driving, then the system can process large amounts of driving data and generate paths, but the system fails to recognize undefined objects and identify intentions in novel driving situations

Engineering Contradiction:
Improvedata processing capabilityVSAvoidrecognition of undefined objects and intentions
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The autonomous driving system is divided into two distinct modules: a data-driven module for processing sensor data and generating initial paths, and a knowledge-driven module for providing guide information on path modifications. This segmentation allows each module to specialize in its strength while compensating for the other's weaknesses.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The knowledge-driven module acts as an intermediary between the data-driven module and the final path determination. It receives the initial path and sensor data, applies knowledge-based reasoning about undefined objects and pedestrian intentions, and outputs guide information that modifies the initial path.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If a knowledge-driven approach with generative neural network is implemented, then the system can respond to new driving situations, but the device complexity increases

Engineering Contradiction:
Improveresponse to new driving situationsVSAvoidsystem architecture complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The knowledge-driven module uses a generative neural network model that can handle multiple types of undefined objects and driving situations through a single unified architecture. This multi-functional approach allows the system to respond to various novel situations without requiring separate specialized modules for each scenario.

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

Solution Approach 2:

The system pre-processes sensor data through encoding before inputting it to the generative neural network model. This preliminary encoding action transforms raw sensor data into a format that the knowledge-driven module can efficiently process, reducing the computational complexity during real-time path modification.

Inventive Principle:
Principle #10Preliminary action

3Speed

If the system determines paths at high frequency, then real-time navigation is achieved, but computational resources are consumed continuously

Engineering Contradiction:
Improvepath determination frequencyVSAvoidcomputational resource consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system determines the initial path at high frequency using the data-driven module, but the knowledge-driven module provides path modifications at a lower frequency. This partial action approach ensures real-time navigation responsiveness while reducing computational resource consumption by applying complex knowledge-based processing only when necessary.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The data-driven module performs preliminary path determination at high frequency, providing a baseline path that is then refined by the knowledge-driven module. This preliminary action allows the system to maintain real-time responsiveness while deferring computationally intensive operations to only when needed for path modification.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4687117A1Method and device with autonomous driving
Publication Date: 2026.02.04 SAMSUNG ELECTRONICS CO LTD
  • EP4687117A1 patent drawingFigure 1
  • EP4687117A1 patent drawingFigure 2
  • EP4687117A1 patent drawingFigure 3

AI summary

A method of determining a final path of a moving object includes acquiring pieces of sensor data, determining a first path of the moving object based on the pieces of sensor data, inputting at least one piece of sensor data among the pieces of sensor data into an encoder that encodes the at least one piece of sensor data, inputting the encoded at least one piece of sensor data into a generative neural network model that generates guide information on a path of the moving object, and determining the final path of the moving object, based on the first path and the guide information.