Neural Network Tracking of Moving Objects with Unknown Equations of Motion
arXiv:2003.08362v1
AI Analysis
This addresses tracking challenges in fields like robotics or surveillance, but appears incremental as it builds on existing filter methods.
The paper tackles the problem of tracking moving objects with unknown motion equations using noisy coordinate measurements, and demonstrates that their neural network method outperforms the Kalman filter in specific scenarios.
In this paper we present a Neural Network design that can be used to track the location of a moving object within a given range based on the object's noisy coordinates measurement. A function commonly performed by the KLMn filter, our goal is to show that our method outperforms the Kalman filter in certain scenarios.