Enhancing vehicle-mountable multiple object tracking systems with embeddable Ising machines
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.This study introduces a novel multiple object tracking system for autonomous vehicles. It uses a quantum-inspired algorithm to solve complex assignment problems, enabling real-time tracking through long occlusions.
Area Of Science
- Robotics and Computer Vision
- Artificial Intelligence
- Quantum Computing Applications
Background
- Multiple object tracking is crucial for autonomous systems like vehicles and robots.
- Current methods struggle with the assignment problem, especially during long-term occlusions.
- Machine learning has improved object similarity but not temporal association assignment.
Purpose Of The Study
- To develop a vehicle-mountable multiple object tracking system.
- To address the challenge of flexible assignment for tracking through occlusions.
- To achieve real-time performance for autonomous applications.
Main Methods
- Formulated the flexible assignment problem as a nondeterministic polynomial-time hard problem.
- Utilized an embedded Ising machine with a quantum-inspired simulated bifurcation algorithm.
- Implemented on a vehicle-mountable computing platform for real-time processing.
Main Results
- Demonstrated a flexible assignment function for tracking through multiple long-term occlusions.
- Achieved real-time system-wide throughput exceeding 20 frames per second.
- Enhanced tracking functionality for autonomous mobile vehicles and robots.
Conclusions
- The developed system effectively solves the complex assignment problem in multiple object tracking.
- Quantum-inspired algorithms on Ising machines offer a viable solution for real-time, robust tracking.
- This technology enhances the capabilities of autonomous systems in dynamic environments.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
