Image-Space 3D Trajectory Planning Around Uncertain Depth Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating through environments with uncertain depth estimates, such as those with complex shapes like trees with intermittent foliage, which can lead to unreliable depth measurements and increased collision risks.
Innovation Solution
The implementation of image space motion planning techniques that identify regions with low confidence depth estimates and associate them with higher collision risks, optimizing the vehicle's trajectory to avoid these areas by projecting the 3D trajectory into the image space and adjusting the path to minimize risk, using cost functions and machine learning to assess danger and uncertainty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If visual odometry is used to estimate position and orientation, then autonomous navigation is enabled, but depth estimation reliability deteriorates in environments with complex shapes like trees with intermittent foliage
Solution Approach 1:
The patent introduces an intermediary processing step between image capture and navigation decision-making: depth uncertainty identification. This intermediary analysis detects regions with unreliable depth estimates (such as those caused by intermittent foliage) and uses this information to adjust navigation planning, preventing the system from making decisions based on unreliable depth data while maintaining autonomous operation
Solution Approach 2:
The patent replaces reliance on purely mechanical/physical depth measurement systems with a computational approach that identifies and flags uncertain depth regions. Instead of trusting all depth measurements equally, the system uses image processing and uncertainty analysis to substitute unreliable mechanical depth data with computationally-derived reliability assessments
2Adaptability or versatility
If the vehicle navigates through areas with complex shapes, then exploration capability is improved, but collision risk increases due to unreliable depth estimates
Solution Approach 1:
The patent applies preliminary action by identifying uncertain depth regions before the vehicle commits to a navigation path. The system analyzes depth uncertainty in advance, marks problematic areas, and incorporates this information into trajectory planning, allowing the vehicle to avoid collision-prone regions while still exploring adaptable paths through reliable areas
Solution Approach 2:
The patent implements feedback by using identified uncertain depth regions to adjust navigation planning. The system continuously monitors depth reliability, feeds this information back into the motion planning algorithm, and modifies the trajectory to avoid areas with high uncertainty, creating a closed-loop system that reduces collision risk while maintaining exploration capability
3Speed
If traditional motion planning is used, then navigation speed is maintained, but safety deteriorates due to inability to account for depth estimation uncertainty
Solution Approach 1:
The patent segments the navigation space into regions of certain and uncertain depth estimates. By dividing the environment into these distinct zones, the system can maintain high-speed navigation through reliable areas while applying caution or avoidance strategies in uncertain regions, thus preserving overall navigation speed while improving safety
Data Source
AI summary
An autonomous vehicle that is equipped with image capture devices can use information gathered from the image capture devices to plan a future three-dimensional (3D) trajectory through a physical environment. To this end, a technique is described for image-space based motion planning. In an embodiment, a planned 3D trajectory is projected into an image-space of an image captured by the autonomous vehicle. The planned 3D trajectory is then optimized according to a cost function derived from information (e.g., depth estimates) in the captured image. The cost function associates higher cost values with identified regions of the captured image that are associated with areas of the physical environment into which travel is risky or otherwise undesirable. The autonomous vehicle is thereby encouraged to avoid these areas while satisfying other motion planning objectives.