Vehicle Object Intent Characterization With Cached Feature Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately predicting the behavior and intent of objects in their environment, particularly pedestrians, due to rapid changes in movement and the need for efficient computational processing to ensure safe navigation.
Innovation Solution
The implementation of a two-model system where a first model processes image data to extract features, and a second model determines attributes by combining these features with cached vectors, allowing for intent prediction with fewer image frames and reduced computational load, thereby improving vehicle safety and operational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple image frames are processed to predict object intent, then prediction accuracy is improved, but computational load and processing time increase
Solution Approach 1:
The system segments the computational task into two distinct models: a first model that extracts features from image frames, and a second model that determines attributes by combining these features with cached vectors. This segmentation allows the system to process fewer image frames while maintaining prediction accuracy, as the feature extraction and attribute determination are performed separately and efficiently.
Solution Approach 2:
The system performs preliminary action by caching feature vectors from previously processed image frames. When a new image frame is received, the second model combines it with cached vectors to determine object attributes, eliminating the need to process all historical frames in full detail. This preliminary caching reduces computational load while preserving prediction accuracy.
2Measurement precision
If multiple image frames are processed to predict object intent, then prediction accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary action by caching feature vectors from previously processed image frames. When a new image frame is received, the second model combines it with cached vectors to determine object attributes, eliminating the need to process all historical frames in full detail. This preliminary caching reduces computational load while preserving prediction accuracy.
Solution Approach 2:
The system extracts only the essential features from image frames using the first model, and then uses these extracted features along with cached vectors in the second model to determine attributes. This extraction approach focuses computational resources on the most relevant information, reducing processing time while maintaining prediction accuracy.
3Device complexity
If fewer image frames are processed, then computational resources are reduced, but prediction accuracy may deteriorate
Solution Approach 1:
The system performs preliminary action by caching feature vectors from previously processed image frames. When a new image frame is received, the second model combines it with cached vectors to determine object attributes, eliminating the need to process all historical frames in full detail. This preliminary caching reduces computational load while preserving prediction accuracy.
Solution Approach 2:
The system uses cached vectors as copies of previously extracted features, allowing the second model to reconstruct object attributes without reprocessing the original historical image frames. This copying mechanism preserves the essential information needed for accurate prediction while significantly reducing computational resource requirements.
Data Source
AI summary
A vehicle computing system may implement techniques to determine attributes (or intent) of an object detected by a vehicle operating in the environment. The techniques may include determining a set of features with respect to a detected object by a first model and determining, by a second model and based on the set of features, one or more attributes of the object. The first model and the second model may be configured to process at least one image frame to determine the one or more attributes of the object. A model may receive sensor data as an input, and output features and/or an attribute for the detected object. Based on the attribute(s) of the object, a vehicle computing system may control operation of the vehicle.