Automated vehicles rely on their environment model, usually generated by a tracking module using sensor data, to make decisions. Therefore, estimating the accuracy of the tracking module is vital for the safe and reliable operation of the vehicle. This work makes a step towards this goal by providing a detection probability estimation method with a self-monitored quality assessment for the labeled multi-Bernoulli filter. We demonstrate the significance of the proposed quality index by comparing it with the actual estimation error calculated with ground truth data. This shows that the developed index is a meaningful value that can be computed online without ground truth data.
Self-Monitored Detection Probability Estimation for the Labeled Multi-Bernoulli Filter
24.09.2024
683932 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Interaction-Aware Labeled Multi-Bernoulli Filter
IEEE | 2023
|British Library Conference Proceedings | 2022
|