Prediction of SLAM ATE Using an Ensemble Learning Regression Model and 1-D Global Pooling of Data Characterization
by Islam Ali, Bingqing (Selina) Wan, and Hong Zhang
Published onMay 28, 2024
Prediction of SLAM ATE Using an Ensemble Learning Regression Model and 1-D Global Pooling of Data Characterization
·
ABSTRACT
Robustness and resilience in simultaneous localization and mapping (SLAM) are critical requirements for modern autonomous robotic systems. One of the essential steps to achieving robustness and resilience is the ability of SLAM to have an integrity measure for its estimates, thus having internal fault tolerance mechanisms to deal with performance degradation. In this work, we introduce a novel method for predicting SLAM localization error based on the characterization of raw sensor inputs. The proposed method relies on using a random forest regression model trained on 1-D global pooled features generated from characterized raw sensor data. The model is validated by using it to predict the performance of ORB-SLAM3 on three different datasets running in four different operating modes, resulting in an average prediction accuracy of up to 93.1 % and 80.45 % for ATE and APE, respectively. Then, the paper studies the quality of prediction with limited training data and proves that we can maintain proper ATE and APE prediction quality when training on only 20 % and 40% of the data, respectively. Finally, the paper discusses the impact of out-of-distribution predictions on prediction accuracy.
Creative Commons Attribution 4.0 International License (CC-BY 4.0)