Pedestrian Crossing Prediction Using Trajectory and Road Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately predicting whether pedestrians will cross a road in the near future due to complexities in scene comprehension and pedestrian-vehicle interactions.
Innovation Solution
A vehicle system integrates multiple sensors and machine learning modules to determine pedestrian characteristics, generate trajectory predictions, and overlay these on road segments to forecast whether a pedestrian will be on or off the road, using a combination of crossing intention, motion state estimation, and vehicle velocity to generate precise road crossing predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensors and machine learning modules are integrated to improve prediction accuracy, then the reliability of pedestrian road crossing prediction is improved, but the device complexity increases
Solution Approach 1:
The system segments the prediction task into distinct functional modules: sensor data acquisition, pedestrian characteristic determination, trajectory prediction, and road crossing prediction. Each module processes specific aspects independently, allowing for improved accuracy through specialized processing while managing complexity through modular architecture.
Solution Approach 2:
The system merges multiple data sources including sensor data, map data, and trajectory predictions into a unified prediction framework. By combining these diverse information streams through integrated processing, the system achieves higher reliability in road crossing predictions than any single source could provide alone.
2Measurement precision
If multiple characteristics and trajectory predictions are analyzed to improve prediction accuracy, then the measurement precision of pedestrian behavior prediction is improved, but the loss of time for processing increases
Solution Approach 1:
The system performs preliminary determination of pedestrian characteristics and generates trajectory predictions in advance of the final road crossing prediction. By pre-processing and analyzing pedestrian behavior patterns, motion states, and intended trajectories beforehand, the system reduces the computational burden during critical decision-making moments, thereby minimizing processing time while maintaining high accuracy.
Solution Approach 2:
The system continuously tracks pedestrian characteristics, motion states, and trajectory predictions over time rather than performing discrete analyses. This continuous monitoring and updating allows the system to maintain accurate predictions with reduced processing intervals, as the system already has current information about pedestrian behavior patterns and intended paths.
Data Source
AI summary
A vehicle system includes at least one sensor configured to capture one or more images of a pedestrian positioned near a road, and a control module in communication with the at least one sensor. The control module is configured to determine one or more characteristics associated with the pedestrian positioned near to the road based on the one or more captured images, generate a trajectory prediction for the pedestrian, overlay the trajectory prediction for the pedestrian on a road segmentation of the road, and generate a road crossing prediction for the pedestrian based on the one or more characteristics and the overlayed trajectory prediction. The road crossing prediction forecasts whether the pedestrian will be on or off the road. Other example vehicle systems and methods for forecasting a future presence of a pedestrian on a road are also disclosed.