Multi-Sensor Fusion State Estimation for MAV Altitude Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Micro-aerial vehicles (MAVs) face challenges in state estimation due to limited payload, power, and computing resources, making autonomous navigation, obstacle avoidance, and environment perception difficult, especially in GPS-denied environments and high-altitude flights, where traditional sensors like GPS and IMUs are unreliable.
Innovation Solution
A novel state estimation technique that fuses long-range stereo visual odometry with GPS, barometric, and inertial measurement unit (IMU) measurements, using a stochastic cloning Extended Kalman Filter (EKF) estimator to generate accurate and smooth state estimates, and switches between stereo and monocular modes for effective low- and high-altitude operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS and IMU are used for state estimation, then positioning information is provided, but GPS cannot provide accurate positioning information for MAVs especially in terms of altitude and IMU integration drifts very fast without global rectification information
Solution Approach 1:
The patent combines multiple sensing systems (GPS, barometer, IMU, and visual sensors) into an integrated state estimation system. The GPS provides horizontal position, the barometer provides altitude, the IMU provides orientation and velocity, and visual sensors provide additional constraints. This merging of sensors compensates for individual sensor deficiencies and improves overall positioning accuracy and reliability.
Solution Approach 2:
The patent introduces visual sensors and barometric altitude as intermediary measurements that help rectify the drift in IMU integration. The visual system provides external reference points for position and orientation correction, while the barometer provides independent altitude measurement that mediates between GPS and IMU altitude estimates, improving the reliability of state estimation.
2Measurement precision
If stereo visual sensors are used for odometry, then six DoF pose measurements are generated, but stereo VO is limited to short range measurements and degenerates to monocular system at high altitude
Solution Approach 1:
The patent implements a dynamic switching mechanism that adapts the visual odometry system's operating mode based on altitude. At low altitudes where stereo baseline-depth ratio is favorable, the system operates in stereo mode for accurate six-DoF pose measurements. At high altitudes where stereo effectiveness degrades, the system dynamically switches to monocular mode with adjusted processing, maintaining adaptability across varying flight conditions.
Solution Approach 2:
The patent changes the operational parameters of the visual odometry system based on altitude conditions. The stereo baseline-depth ratio, a critical parameter, determines the operating mode. By monitoring altitude and adjusting the odometry algorithm parameters accordingly (stereo vs. monocular mode), the system maintains measurement precision across different flight heights.
3Weight of moving object
If light-weight compact sensors are integrated for MAV applications, then payload, power and computing resources are constrained, but state estimation, environment perception and obstacle avoidance become more difficult
Solution Approach 1:
The patent designs a multi-functional sensor system where each sensor serves multiple purposes. The IMU provides not only orientation but also velocity estimates and serves as a reference for visual odometry. The visual sensors provide both navigation information and environmental perception data. The barometer provides altitude measurement and helps correct vertical drift. This multi-functionality reduces the need for separate dedicated sensors, managing payload constraints while achieving complex state estimation.
Data Source
AI summary
A state estimation system that utilizes long-range stereo visual odometry that can degrade to a monocular system at high-altitude, and integrates GPS, Barometer and IMU measurements. The system has two main parts: An EKF that is loosely fused and a long-range visual odometry part. For visual odometry, the system takes the EKF information for robust camera pose tracking, and the visual odometry outputs will be the measurement for EKF state update.