Speed estimation from Doppler radar gets confidence aware and robust
Confidence-Aware Speed-over-Ground Estimation from Doppler Spectra via Reconstruction-Based Representation Learning
Machine Learning
Summary
Estimating how fast a train or vehicle moves using radar signals is important but can be tricky when conditions are not good. The authors propose a new method that uses machine learning to estimate the speed and how confident the system is about that estimate from radar data. They train their system specifically with railway data and add a way to measure reliability for each speed prediction. Their experiments show better speed accuracy than traditional methods, especially in difficult conditions, while also knowing when predictions might be wrong.
What this means in practice
- •For railway system engineers: Improve train speed monitoring by integrating confidence-aware Doppler radar speed estimates to increase safety and accuracy under adverse signal conditions.
- •For mobile robotics developers: Incorporate a robust speed estimation method with confidence indicators into ground vehicles navigating challenging environments using Doppler radar data.
Authors
Malte Stelzer, Marvin Sach, Timo Bartels, Dinu Mihailescu-Stoica, Sören Kohnert, Florian Poprawa, Marc Zeller, Tim Fingscheidt
Abstract
Reliable speed-over-ground estimation from ground-facing Doppler spectra is essential for mobile platforms such as railway systems. Classical signal-processing approaches can provide accurate estimates under favorable conditions, but often degrade in challenging conditions. We propose confidence-aware speed-over-ground estimation trained on rail-domain data that estimates speed over ground and frame-level confidence from Doppler spectra with several novel contributions: First, we transfer the idea of exploiting a reconstruction task by an autoencoder for performance estimation from previous work to Doppler RaDAR. Second, we append a speed-over-ground estimation head operating on the autoencoder's latent space features. Third, we omit the reconstruction decoder during inference but add a low-cost dedicated confidence head for frame-wise error-bound estimation. Fourth, we reveal how noise augmentation training can provide high robustness to the method. Experiments show that our approach improves speed estimation accuracy over classical baselines while providing informative confidence estimates that indicate the reliability of individual predictions, even under adverse signal conditions.