Enhancing vehicle-mountable multiple object tracking systems with embeddable Ising machines

The cognitive function of tracking multiple objects, essential for autonomous mobile vehicles or autonomous robots, involves object detection and their temporal associations. While significant progress has recently been made in machine learning to elaborate the similarity matrix between the objects that have been recognized and the objects detected in the current video frame, less progress has been made on the assignment problem that ultimately determines temporal associations, which is a combinatorial optimization problem. Here we show a vehicle-mountable multiple object tracking system with a flexible assignment function for tracking through multiple long-term occlusion events. To solve the flexible assignment problem, formulated as a nondeterministic polynomial-time hard problem, the system relies on an embedded Ising machine based on a quantum-inspired algorithm called simulated bifurcation. Using a vehicle-mountable computing platform, we demonstrate real-time system-wide throughput of more than 20 frames per second with the enhanced functionality.