Trajectory Similarity Scoring via Deep Neural Network

The patent introduces a deep neural network model as an intermediary between trajectory data and evaluation results. The DNN model processes trajectory features and generates similarity scores, acting as a mediator that transforms complex trajectory comparison into a standardized evaluation metric. This resolves the contradiction by providing high measurement precision through the DNN intermediary while keeping the overall system architecture manageable.

The patent transforms trajectory evaluation from direct spatial-temporal comparison into a parameter-based similarity scoring system. By extracting key features from trajectories and evaluating them through DNN-generated similarity scores, the system changes the evaluation parameters from raw coordinate data to processed similarity metrics, achieving higher measurement precision without proportionally increasing system complexity.

The patent extracts only the most relevant features from selected trajectory points rather than processing all available data. By selectively extracting key characteristics that most influence trajectory similarity, the system achieves high evaluation accuracy while reducing the computational burden and time required for processing.

The patent applies partial action by selecting specific trajectory points for feature extraction rather than analyzing every point along the trajectory. This selective approach provides sufficient evaluation accuracy for most practical purposes while significantly reducing computational time compared to exhaustive analysis of all trajectory data.