Autonomous Driving Path Planning Using Parallel Processing Threads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in navigating through environments with unpredictable moving obstacles, such as pedestrians changing direction, as conventional trajectory planning techniques rely heavily on stable reference lines and struggle to accurately predict the paths of such obstacles to avoid collisions.
Innovation Solution
The system determines a predicted area where moving obstacles may move, allowing the autonomous vehicle to plan a path that avoids this area, thereby increasing the probability of collision avoidance by using a perception and planning system that includes modules for localization, perception, prediction, decision-making, and path planning, incorporating sensors like cameras, radar, and LIDAR to detect and predict the movements of obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional trajectory planning techniques rely on stable reference lines, then the system can generate stable trajectories, but it cannot accurately predict paths of unpredictable moving obstacles
Solution Approach 1:
The system segments the trajectory planning into two independent modules: reference line generation (providing stable baseline) and obstacle prediction (handling dynamic objects). This segmentation allows each module to specialize - the reference line module ensures stability while the prediction module uses machine learning to accurately track moving obstacles, resolving the contradiction between stability and adaptability.
Solution Approach 2:
The system introduces an intermediary prediction module that acts as a mediator between the stable reference lines and the dynamic obstacle environment. This intermediary uses sensor data and machine learning models to predict obstacle trajectories, allowing the vehicle to maintain stable reference line following while adapting to unpredictable moving obstacles through the intermediary's predictions.
2Reliability
If the system uses multiple sensors and prediction modules to accurately predict obstacle movements, then collision avoidance improves, but system complexity increases
Solution Approach 1:
The system merges multiple sensors (cameras, LIDAR, radar) into a unified perception framework that feeds into integrated prediction modules. By combining sensor inputs and prediction functions into cohesive modules rather than separate systems, the architecture achieves high collision avoidance capability while managing complexity through integration and shared processing pipelines.
Solution Approach 2:
The prediction modules are designed with multi-functionality, serving both obstacle detection and trajectory prediction purposes simultaneously. This universal design allows the same computational infrastructure to handle multiple safety-critical functions, reducing overall system complexity while maintaining high reliability for collision avoidance.
3Speed
If the autonomous vehicle uses parallel processing with multiple threads for path determination, then processing speed improves, but computational resource consumption increases
Solution Approach 1:
The system implements periodic action by dividing path determination into discrete planning cycles that execute at optimized intervals. Instead of continuous computation, the parallel processing threads are activated periodically to perform path determination, reference line generation, and obstacle prediction in batches. This approach increases processing speed through parallel execution while reducing energy consumption by avoiding constant computational activity.
Solution Approach 2:
The system applies dynamics by adaptively adjusting the number and activation of parallel threads based on operational context. During high-risk situations with multiple moving obstacles, more threads are activated to increase processing speed. During stable cruising conditions, fewer threads are active, reducing computational energy consumption while maintaining adequate path determination capability.
Data Source
AI summary
An autonomous driving vehicle (ADV) may determine a predicted path for a moving obstacle and speeds for different portions of the path. The ADV use multiple threads in parallel to determine the path and speeds for the different portions of the path.