Visual-Inertial Mapping with Hybrid Point Grid Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for providing fast, accurate, and reliable positional awareness to robots and wearable devices, such as those used in virtual reality and augmented reality, face challenges including high costs, power consumption issues, and limitations in recognizing locations and obstructions quickly, with existing methods like RFID/WiFi, depth sensors, and visual approaches falling short in terms of speed and accuracy.
Innovation Solution
The implementation of a visual-inertial sensor system that combines grayscale and colored cameras with a multi-axis inertial measurement unit (IMU) to process image and inertial data efficiently, offloading computational tasks to a low-power sensor module, and using a hybrid point grid for localization and mapping, enabling cost-effective and energy-efficient positional awareness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If RFID/WiFi approaches are used for positional awareness, then coverage area is improved, but measurement precision and speed deteriorate
Solution Approach 1:
The patent combines multiple sensing modalities (visual sensors, depth sensors, inertial sensors, RFID/WiFi) into a unified sensor system that processes data from all sources simultaneously. This fusion approach allows the system to maintain wide coverage from RFID/WiFi while achieving high precision through visual and depth sensor data correlation.
Solution Approach 2:
The patent introduces markers as intermediary objects that mediate between the broad coverage of RFID/WiFi and the precision requirements of visual positioning. These markers provide visual features for accurate localization while being placed within the broader RFID/WiFi coverage area, enabling the system to achieve both goals.
2Measurement precision
If depth sensor based approaches are used, then measurement precision is improved, but use of energy and cost increase
Solution Approach 1:
The patent implements periodic or event-triggered depth sensing rather than continuous operation. The depth sensors are activated only when needed for specific tasks or when visual-inertial tracking requires calibration, significantly reducing power consumption while maintaining measurement precision when required.
Solution Approach 2:
The patent makes the depth sensor serve multiple functions: primary depth measurement, visual-inertial odometry calibration, obstacle detection, and environment mapping. This multi-functionality justifies the energy cost by extracting maximum utility from each activation, reducing the frequency of operation.
3Ease of manufacture
If visual approaches are used, then cost is reduced, but speed and reliability deteriorate in fast motion applications
Solution Approach 1:
The patent merges visual sensing with inertial sensing to create a visual-inertial system. The inertial sensors (accelerometers, gyroscopes) provide high-speed motion data that complements the lower-speed visual data, enabling the system to maintain accuracy during fast motion while using cost-effective visual sensors.
Solution Approach 2:
The patent implements feedback loops where inertial sensor data continuously corrects and updates the visual tracking estimates in real-time. This feedback mechanism allows the system to compensate for visual processing delays and maintain high-speed tracking performance with affordable visual sensors.
4Ease of manufacture
If visual approaches are used, then cost is reduced, but measurement precision deteriorates due to scale ambiguity
Solution Approach 1:
The patent introduces markers as intermediary objects that provide known geometric features and scale references. These markers serve as mediators between the low-cost visual sensors and the need for accurate scale estimation, eliminating scale ambiguity by providing visual features with known dimensions and configurations.
Solution Approach 2:
The patent changes the parameter space by incorporating inertial sensor measurements (acceleration, velocity, orientation) alongside visual features. This multi-parameter approach allows the system to disambiguate scale and position by correlating visual observations with inertial data, achieving accurate positioning with cost-effective visual sensors.
Data Source
AI summary
The described positional awareness techniques employing visual-inertial sensory data gathering and analysis hardware with reference to specific example implementations implement improvements in the use of sensors, techniques and hardware design that can enable specific embodiments to provide positional awareness to machines with improved speed and accuracy.