Visual-Inertial Sensor Fusion for Robust Positional Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for providing fast, accurate, and reliable positional awareness to robots and wearable devices face challenges such as high costs, power consumption issues, and limitations in recognizing locations and obstructions quickly, with existing methods like RFID/WiFi, depth sensors, and visual approaches being expensive, power-intensive, or slow.
Innovation Solution
The implementation of a visual-inertial sensory system that combines stereo imaging sensors with a multi-axis inertial measurement unit (IMU) to estimate changes in the environment, using low-cost grayscale and RGB cameras, and offloading computational tasks to a low-power sensor module to reduce energy consumption and costs, while maintaining accurate tracking and mapping.
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 cost and accuracy are worsened
Solution Approach 1:
The patent combines multiple sensing modalities (visual sensors, depth sensors, inertial sensors, RFID, WiFi) into a unified sensor fusion system. This merging allows the system to leverage the broad coverage of RFID/WiFi while compensating for their accuracy limitations through complementary data from visual and depth sensors, achieving both wide coverage and high precision simultaneously.
Solution Approach 2:
The system uses a composite sensing architecture that integrates different types of sensors (optical, electromagnetic, inertial) with complementary strengths. Each sensor type contributes its unique capabilities to the overall positional awareness system, creating a robust multi-modal sensing solution that overcomes the limitations of individual sensor types.
2Measurement precision
If depth sensor based approaches are used, then positional accuracy is improved, but power consumption and cost are worsened
Solution Approach 1:
The system implements periodic or event-triggered depth sensing rather than continuous operation. Depth sensors are activated only when needed (e.g., during initialization, when visual tracking is uncertain, or at specific intervals), allowing the system to maintain positional accuracy when required while minimizing power consumption during normal operation.
Solution Approach 2:
The visual-inertial odometry system serves itself by using low-power visual and inertial sensors for continuous tracking, reserving depth sensors for occasional calibration or correction. The system autonomously determines when depth sensing is necessary based on tracking confidence metrics, eliminating the need for continuous high-power depth sensing.
3Ease of manufacture
If visual approaches are used, then cost is reduced, but speed and reliability are worsened
Solution Approach 1:
The patent merges visual sensing with inertial sensing to create a visual-inertial odometry system. The inertial sensors (accelerometers, gyroscopes) provide high-speed motion data that compensates for the slower processing of visual approaches, while visual data provides cost-effective environmental context. This combination achieves both speed and cost-effectiveness.
Solution Approach 2:
The system replaces purely computational visual processing with a hybrid approach that incorporates mechanical inertial sensors. The inertial measurement unit (IMU) provides direct physical measurements of acceleration and orientation that are computationally efficient and high-speed, substituting for some of the heavy computational lifting that would otherwise be required by visual methods alone.
4Measurement precision
If marker based approaches are used, then positional accuracy is improved, but adaptability is worsened
Solution Approach 1:
The system extracts and removes the requirement for artificial markers from the environmental assumptions. By using natural visual features (corners, edges, textures) and inertial data instead of markers, the system achieves markerless operation that works in diverse environments without requiring pre-placed markers, thereby improving adaptability while maintaining accuracy through alternative feature detection methods.
Solution Approach 2:
The visual-inertial system creates a universal positioning solution that works across multiple environments (indoor, outdoor, structured, unstructured) without requiring environment-specific configurations. The system universally detects and uses natural visual features combined with inertial data, making it adaptable to any environment where visual features exist, eliminating the need for marker-based systems that are environment-specific.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables efficient and cost-effective positional awareness for robots and wearable devices, improving speed and accuracy while reducing energy consumption and costs, enabling reliable tracking and mapping in various applications including robotics, VR, and AR.
Implementation Method 1
a multi-axis inertial measurement unit (IMU) to estimate changes in the environment
Implementation Method 2
stereo imaging sensors with a multi-axis inertial measurement unit
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.