UAV Object Tracking With Sensor Fusion for Dynamic Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle navigation systems, particularly those using visual sensors, face challenges in accurately tracking objects in dynamic environments and maintaining robust motion planning due to limitations in image processing and sensor integration.
Innovation Solution
An unmanned aerial vehicle (UAV) equipped with multiple image capture devices and a hybrid mechanical-digital gimbal system for adjusting image capture orientation, combined with a navigation system that includes motion planning and tracking subsystems, fuses visual data with inertial and GPS information to predict object trajectories and maintain object tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual sensors are used for autonomous vehicle navigation, then the system can estimate position and orientation, but measurement precision deteriorates in dynamic environments with moving objects
Solution Approach 1:
The patent introduces an object tracking subsystem as an intermediary between the visual sensors and the navigation system. This tracker maintains dedicated object tracks with predicted positions, serving as a mediator that provides stable reference points even when visual conditions are poor. The tracker uses motion models and prediction algorithms to maintain reliable object position estimates independent of immediate visual sensor quality.
Solution Approach 2:
The system performs preliminary action by predicting object positions and maintaining tracking states before visual confirmation is obtained. The object tracker continuously updates predicted positions using motion models, so when visual sensors detect an object, the system already has a prepared track and expected position, enabling faster and more accurate association and measurement.
2Measurement precision
If multiple image capture devices are used for stereoscopic capture, then depth computation accuracy improves, but device complexity increases
Solution Approach 1:
The patent merges the functionality of multiple image capture devices into a unified stereoscopic vision system. The navigation system integrates images from multiple cameras, combining their data to compute depth and three-dimensional object positions. This merging approach achieves accurate depth measurement while managing system complexity through integrated processing rather than treating each device separately.
3Reliability
If visual data is fused with inertial and GPS information, then navigation robustness improves, but processing complexity increases
Solution Approach 1:
The navigation system is designed with multi-functionality to handle multiple sensor types (visual sensors, inertial sensors, GPS) through a unified processing framework. The system performs visual odometry, object tracking, and multi-sensor fusion using the same core navigation algorithms, achieving robust navigation without requiring separate specialized processing paths for each sensor type.
Solution Approach 2:
The system implements feedback mechanisms where the object tracker provides predicted positions that are compared with actual sensor measurements. This feedback loop allows the system to continuously refine its estimates and compensate for individual sensor errors, improving overall navigation robustness while using a coordinated multi-sensor approach that manages processing complexity through iterative refinement.
Data Source
AI summary
Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.