Pedestrian Motion Prediction Using Context-Semantic Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object motion prediction methods struggle to incorporate context information with motion data from different domains, leading to reduced accuracy and efficiency, particularly in predicting pedestrian behavior, which is unpredictable and difficult to anticipate using existing techniques.
Innovation Solution
A method and device that generate a semantic map with context information, create a motion flow map, and combine these into a motion-semantic map to predict object motion, aligning dimensions for accurate pedestrian motion prediction and path planning in autonomous vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional object motion prediction methods use only location data and sensor tracking, then the system complexity remains low, but the prediction accuracy deteriorates especially for pedestrian behavior
Solution Approach 1:
The patent merges multiple data domains (location data, sensor tracking data, and context information) into a unified motion prediction model. By combining these previously separate data sources, the system achieves higher prediction accuracy for pedestrian behavior while managing complexity through integrated processing.
Solution Approach 2:
The patent introduces a context information processing module as an intermediary that bridges location data and sensor tracking data. This intermediary component extracts and integrates contextual factors (such as environmental conditions, pedestrian attributes, and situational context) to enhance prediction accuracy without directly complicating the core motion prediction engine.
2Measurement precision
If context information is integrated with motion data from different domains, then the prediction accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent segments the data processing into distinct modules: one for handling location data, another for sensor tracking data, and a separate context information processing module. Each module processes its specific data type independently, then the results are integrated. This segmentation reduces overall data processing complexity by avoiding the need to process all data types simultaneously in a single complex pipeline.
Solution Approach 2:
The patent creates a universal data integration framework that can handle multiple data domains (location, sensor tracking, context information) through a common processing architecture. This multi-functional system uses standardized interfaces and processing routines that work across different data types, reducing the complexity that would otherwise arise from domain-specific processing for each data type.
3Ease of manufacture
If conventional methods use separate encoding for different data domains, then the implementation remains simple, but the information compatibility deteriorates
Solution Approach 1:
The patent merges previously separate encoding schemes into a unified representation framework. By combining location data, sensor tracking data, and context information into a single integrated data structure, the system eliminates information compatibility issues that arose from separate encodings while maintaining implementation simplicity through consistent processing rules.
Solution Approach 2:
The patent creates a composite data representation that integrates multiple data domains into a unified structure. Similar to how composite materials combine different materials to achieve superior properties, this composite data structure combines location, sensor, and context information to achieve complete information compatibility while preserving the unique characteristics of each data type.
Data Source
AI summary
The present disclosure relates to a method and device for predicting object motion based on context recognition. Additionally, the present disclosure relates to a method for establishing a moving object path plan based on pedestrian motion prediction in a moving object capable of autonomous driving, using the method for predicting object motion based on context recognition. The method for predicting object motion based on context recognition, according to the present disclosure, includes generating a semantic map for context information associated with an object, generating a motion flow map that includes the motion flow for each object, generating a motion-semantic map based on the semantic and motion flow maps, and performing motion prediction of at least one object based on the motion-semantic map.