Robust Rail-Track Section Identification With Multiple Structured Light Sensors and Kernel-Based Belief Sensor-Credibility Evaluation

Current single-structured light sensor-based rail-track section identification lacks robustness against unstable signal transmission. The decision-level sensor fusion based on evidence theory is fragile to sensor conflict. Aiming at the listed challenges, this study proposes a robust scheme that combines multistructured light sensors and a new evidence-theoretic sensor fusion on the strength of kernel method. In this scheme, multiple sensors are involved to tackle the lack of robustness, and a kernel-induced belief metric (KIBM), which first connects the reproducing kernel Hilbert space (RKHS) representation to belief distance-dominated sensor credibility evaluation, is newly constructed to address sensor conflict in evidence fusion. In addition, the gap between rail-track section identification and evidence theory is filled. Experiments reveal the efficiency and robustness of involving multisensors and the new evidence-theoretic sensor fusion in the identification on rail-track sections.