Abstract:Vision-based estimation of the motion of a moving target is usually formulated as a bearing-only estimation problem where the visual measurement is modeled as a bearing vector. Although the bearing-only approach has been studied for decades, a fundamental limitation of this approach is that it requires extra lateral motion of the observer to enhance the target's observability. Unfortunately, the extra lateral motion conflicts with the desired motion of the observer in many tasks. It is well-known that, once a target has been detected in an image, a bounding box that surrounds the target can be obtained. Surprisingly, this common visual measurement especially its size information has not been well explored up to now. In this paper, we propose a new bearing-angle approach to estimate the motion of a target by modeling its image bounding box as bearing-angle measurements. Both theoretical analysis and experimental results show that this approach can significantly enhance the observability without relying on additional lateral motion of the observer. The benefit of the bearing-angle approach comes with no additional cost because a bounding box is a standard output of object detection algorithms. The approach simply exploits the information that has not been fully exploited in the past. No additional sensing devices or special detection algorithms are required.

Active Target Tracking Using Bearing-only Measurements With Gaussian Process Learning

Irregular Extended Target Tracking with Unknown Measurement Noise Covariance

Learning the Dynamics of Flapping Wing Robots via Gaussian Process Regression during Three-Dimensional Trajectory Tracking.

A Gaussian Process Regression based Dynamical Models Learning Algorithm for Target Tracking

Bearing-Only Formation Tracking Control for Multi-Agent Systems with Time-Varying Velocity Leaders

Adaptive Model Prediction Control-Based Multi-Terrain Trajectory Tracking Framework for Mobile Spherical Robots

An Improved Gaussian Mixture Particle Filter Based Targets Tracking Algorithm for Bearing-Only Tracking System

An Adaptive Target Tracking Algorithm Based on EKF for AUV with Unknown Non-Gaussian Process Noise

Bearings-Only Target Tracking with an Unbiased Pseudo-Linear Kalman Filter

Episodic Gaussian Process-Based Learning Control with Vanishing Tracking Errors

Learning-Based Trajectory Tracking and Balance Control for Bicycle Robots with a Pendulum: A Gaussian Process Approach

An Active-Sensing Approach for Bearing-based Target Localization

Attitude Takeover Control for Noncooperative Space Targets Based on Gaussian Processes with Online Model Learning

Online learning-based trajectory tracking for underactuated vehicles with uncertain dynamics

Collaborative penalized Gaussian mixture PHD tracker for close target tracking.

Real-Time Adaptive Safety-Critical Control with Gaussian Processes in High-Order Uncertain Models

Physics-informed Gaussian Processes as Linear Model Predictive Controller

Model Predictive Tracking Control for Three-wheeled Mobile Robots Combined with Gaussian Process Regression

A Bearing-Angle Approach for Unknown Target Motion Analysis Based on Visual Measurements

Visual Tracking Via Locally Structured Gaussian Process Regression