Autonomous Vehicle Path Planning Using Predicted Obstacle Areas

Resolve Bottlenecks,
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 and avoid these unpredictable movements.

Innovation Solution

The system determines a predicted area where a moving obstacle may move, using sensors and machine learning algorithms to calculate possible locations with high probability, allowing the autonomous vehicle to plan a path that avoids this area and increase the likelihood of collision avoidance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional trajectory planning techniques rely on stable reference lines, then the vehicle can generate stable trajectories, but the system cannot accurately predict and avoid unpredictable moving obstacles

Engineering Contradiction:

Improvetrajectory stabilityVSAvoidability to handle unpredictable obstacles

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system transitions from static reference line-based planning to dynamic prediction areas that adapt to moving obstacles. The prediction area is continuously updated based on obstacle detection and movement prediction, allowing the trajectory planner to dynamically adjust avoidance paths while maintaining stability through systematic recalibration.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system adds a temporal dimension to obstacle representation by creating prediction areas that encompass future possible positions. Instead of reacting to current obstacle positions only, the planner works with expanded spatial-temporal zones that predict where obstacles may move, enabling proactive path planning in four-dimensional space-time.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Reliability

If the system creates a prediction area to avoid moving obstacles, then collision avoidance probability increases, but the path planning complexity increases

Engineering Contradiction:

Improvecollision avoidance probabilityVSAvoidpath planning complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The prediction area is segmented into discrete regions based on obstacle type, detection confidence levels, and predicted movement patterns. This segmentation allows the path planner to process different areas with appropriate complexity levels, focusing computational resources on high-risk zones while simplifying low-risk regions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary prediction of obstacle movement areas before actual path planning occurs. By pre-calculating prediction zones and potential conflict areas, the system reduces real-time planning complexity, as the avoidance path generation works with preprocessed spatial constraints rather than raw sensor data.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10800408B2Determining driving paths for autonomous driving that avoid moving obstacles

Publication Date: 2020.10.13 BAIDU USA LLC

AI summary

An ADV may determine a predicted path for a moving obstacle. The ADV may determine a predicted area based on the predicted path. The ADV may determine a path for the ADV based on the predicted area. The path for the ADV may avoid the predicted area when determining the path for the ADV.