Sparse Voxel Path Planning for Low-Latency AR Rendering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing computing systems struggle to efficiently process large 3D data sets for augmented, virtual, and mixed reality applications due to memory and processing constraints, leading to latency issues that can cause user discomfort.

Innovation Solution

Implementing a Sparse Sexaquaternary Tree (SST) volumetric data structure that tags voxels as occupied or empty, allowing for reduced storage and faster processing, and using hardware accelerators to update 3D renderers in real time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If dense volumetric data structures are used to represent 3D environments, then complete spatial information is available for accurate rendering and collision detection, but memory consumption and processing time increase significantly

Engineering Contradiction:

Improvespatial information accuracyVSAvoidmemory consumption

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating between occupied and empty voxels in the volumetric data structure. Only regions containing actual geometry or objects are stored with detailed information, while empty spaces are represented more compactly. This allows the system to maintain high spatial information accuracy where needed while reducing overall memory consumption by avoiding redundant storage of empty regions.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The volumetric space is segmented into discrete voxels that are individually tagged as occupied or empty. This segmentation enables the system to process and store only the necessary portions of the 3D environment, dividing the large-scale spatial data into manageable units that can be efficiently managed in memory and processed in parallel by hardware accelerators.

Inventive Principle:

Principle #1Segmentation

2Reliability

If complete 3D volumetric data is processed in real-time for AR/VR rendering, then high-quality visual output is achieved, but processing latency increases causing user discomfort

Engineering Contradiction:

Improverendering qualityVSAvoidprocessing latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs partial action by processing only the occupied voxels that contain actual geometry or objects, rather than processing the entire volumetric dataset. This selective processing approach maintains rendering quality for visible elements while significantly reducing the computational burden and processing latency that would otherwise cause user discomfort in AR/VR applications.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent replaces traditional software-based volumetric processing with specialized hardware accelerators that are optimized for parallel processing of voxel data. This mechanical substitution enables real-time rendering of complete 3D volumetric data with high quality output while maintaining low latency, as the hardware architecture is specifically designed to handle the computational demands of AR/VR rendering pipelines.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If separate GPU and computer vision subsystems are used in parallel, then specialized processing capabilities are maintained for both graphics and vision tasks, but system complexity and inter-subsystem synchronization overhead increase

Engineering Contradiction:

Improveprocessing capabilityVSAvoidsystem architecture

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges the computer vision subsystem and graphics rendering pipeline by having both subsystems operate on the same volumetric data structure (the voxel grid). Instead of maintaining separate data representations and processing pipelines, the system combines them into a unified architecture where the voxel grid serves as a common data foundation, reducing system complexity while preserving the specialized processing capabilities of both vision and graphics subsystems.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20250327672A1Path planning using sparse volumetric data

Publication Date: 2025.10.23 MOVIDIUS LTD

AI summary

A view of geometry captured in image data generated by an imaging sensor is compared with a description of the geometry in a volumetric data structure. The volumetric data structure describes the volume at a plurality of levels of detail and includes entries describing voxels defining subvolumes of the volume at multiple levels of detail. The volumetric data structure includes a first entry to describe voxels at a lowest one of the levels of detail and further includes a number of second entries to describe voxels at a higher, second level of detail, the voxels at the second level of detail representing subvolumes of the voxels at the first level of detail. Each of these entries include bits to indicate whether a corresponding one of the voxels is at least partially occupied with the geometry. One or more of these entries are used in the comparison with the image data.