Vehicle Trajectory Prediction Using Interaction Scenario Simulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional rule-based planning and control algorithms in autonomous driving systems face challenges in effectively handling complex traffic scenario interactions and iterations due to the exponential expansion of driving scenarios.

Innovation Solution

A method for predicting a vehicle trajectory that involves obtaining scenario interaction results from static and traffic agent perceptions, performing interaction scenario simulation based on an ego vehicle's decision, and determining an optimal driving trajectory using a learned interaction value network to minimize the difference with human driving behavior.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If rule-based planning and control algorithms are used, then the system can handle basic driving scenarios, but it faces difficulties in iteration when new scenario interactions emerge due to exponential expansion of driving scenarios

Engineering Contradiction:
Improveability to handle new scenario interactionsVSAvoidcomplexity of planning and control algorithm
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces rule-based mechanical planning algorithms with a neural network-based learning system. The PNC neural network is trained offline using imitation learning from human driving data, enabling the system to adapt to new scenarios through data-driven patterns rather than exhaustive rule iteration. This substitution allows the system to handle exponentially expanding driving scenarios without proportionally increasing algorithmic complexity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary action by pre-training the PNC neural network offline using large datasets of human driving behavior. This offline training phase prepares the model in advance to handle diverse scenarios, so that during actual autonomous driving operation, the system can quickly adapt to new scenarios without requiring real-time iterative rule development. The pre-learned policies enable rapid deployment to new scenario types.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If complex scenario interactions are considered in trajectory prediction, then the accuracy of driving decisions improves, but the computational complexity and processing time increase

Engineering Contradiction:
Improveaccuracy of trajectory predictionVSAvoidcomputational processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the trajectory prediction task into distinct neural network components: an interaction scenario simulation module that models traffic agent behaviors, and a PNC module that integrates these simulations to predict the ego vehicle's optimal trajectory. This segmentation allows each component to specialize in specific aspects of scenario understanding, improving overall prediction accuracy while maintaining computational efficiency through modular processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system uses copying by simulating multiple possible interaction scenarios between the ego vehicle and traffic agents, then selecting the most probable scenario based on learned patterns from training data. Rather than computing all possible interactions, the system copies realistic scenario patterns from the training dataset, enabling accurate trajectory prediction without exhaustive computational search of all potential interactions.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4527708B1Method for obtaining vehicle trajectory, control apparatus, readable storage medium, and vehicle
Publication Date: 2026.03.25 ANHUI NIO AUTONOMOUS DRIVING TECH CO LTD
  • EP4527708B1 patent drawingFigure 1~2
  • EP4527708B1 patent drawingFigure 3~4
  • EP4527708B1 patent drawingFigure 5

AI summary

The disclosure relates to the technical field of autonomous driving, and provides a method for predicting a vehicle trajectory, a control apparatus, a readable storage medium, and a vehicle, to solve the problem of how to effectively deal with existing highly complex traffic scenario interactions and difficulties in iterative planning and control processes. In the method, scenario interaction results of a traffic agent are obtained based on a static environment perception result and a traffic agent perception result of a vehicle at a current moment, an initial interaction scenario is obtained based on the scenario interaction results, interaction scenario simulation is performed based on the initial interaction scenario and an ego vehicle driving decision to determine an optimal interaction scenario evolution feature of the traffic agent, and an optimal driving trajectory of the vehicle is obtained based on the optimal interaction scenario evolution feature and a state of the vehicle. An optimal interaction scenario evolution process is selected from numerous interaction scenarios by comprehensively considering impact of the ego vehicle driving decision on an environment of the vehicle, thereby obtaining a more accurate optimal driving trajectory.