Mobile Platform Dynamic Object Tracking Using TOF and RANSAC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional SLAM technologies perform poorly in dynamic environments with moving objects, as they are designed for static environments and struggle to accurately detect and track objects, affecting the route design performance of mobile platforms.
Innovation Solution
A method for a mobile platform to detect and track moving objects by acquiring a 3D image using a TOF sensor, removing the floor plane with a RANSAC algorithm, separating objects with a blob algorithm, and estimating their movement using a joint probability data association filter (JPDAF), while registering and tracking dynamic objects to design an obstacle-avoiding route.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional SLAM technology is used in static environments, then localization accuracy is improved, but performance deteriorates in dynamic environments with moving objects
Solution Approach 1:
The patent segments the environment into static and dynamic components by detecting moving objects and separating them from the map data. This allows the SLAM system to process static structures for localization while excluding dynamic objects that would otherwise degrade performance.
Solution Approach 2:
The system performs preliminary detection and classification of moving objects before they interfere with the SLAM localization process. By identifying and removing dynamic object data in advance, the system maintains accurate localization without being affected by subsequent movements.
2Loss of information
If conventional SLAM algorithms process all detected objects, then comprehensive environmental mapping is achieved, but detection precision deteriorates due to interference from moving objects
Solution Approach 1:
The patent extracts and removes data corresponding to moving objects from the overall environmental map. This separation allows the system to maintain a clean static environment model for accurate SLAM while still tracking dynamic objects independently through dedicated detection algorithms.
Solution Approach 2:
The system dynamically adjusts its processing by identifying objects with significant movement and applying different handling strategies: static objects are integrated into the SLAM map while moving objects are excluded from localization calculations but tracked separately for safety and navigation purposes.
3Productivity
If the mobile platform navigates without object tracking, then route design speed is improved, but collision risk increases in dynamic environments
Solution Approach 1:
The system performs preliminary tracking and prediction of moving object trajectories before the mobile platform reaches their potential paths. This advance awareness allows the route planning algorithm to design collision-free routes without real-time reaction delays, maintaining both speed and safety.
Solution Approach 2:
The system continuously monitors moving objects and provides feedback to the route design module. When a moving object's predicted path intersects with the planned route, the system adjusts the route in real-time while maintaining overall navigation efficiency through optimized replanning algorithms.
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 enhances the performance of SLAM in dynamic environments, enabling accurate detection and tracking of moving objects and improving route design for mobile platforms to navigate autonomously without collisions.
Implementation Method 1
a depth map acquired by a time-of-flight (TOF) sensor
Data Source
AI summary
Disclosed herein is a computer-readable medium and method of a mobile platform detecting and tracking dynamic objects in an environment having the dynamic objects. The mobile platform acquires a three-dimensional (3D) image using a time-of-flight (TOF) sensor, removes a floor plane from the acquired 3D image using a random sample consensus (RANSAC) algorithm, and individually separates objects from the 3D image. Movement of the respective separated objects is estimated using a joint probability data association filter (JPDAF).