Resource-aware Hybrid Estimation Framework for Multi-robot Cooperative Localization

In this paper, we present a hybrid estimation framework (HEF) for multi-robot Cooperative Localization (CL) under time-varying, communication-bandwidth constraints. In the presence of severe communication constraints, robots can communicate only a quantized version of their analog measurements to the team. The HEF enables robots to process all available information, i.e., local analog measurements (recorded by its own sensors) and remote quantized measurements (collected and communicated by other robots), to improve the estimation accuracy of CL. However, the existing Hybrid Sign-of-Innovation Kalman filter (H-SOI-KF) is designed to process remote measurements that are quantized to a single bit. Since the communication-bandwidth available for CL is time-varying, the HEF should be resource-aware and utilize additional bandwidth, whenever available, to further improve the estimation accuracy. To achieve this, in this paper, we present two filters, the Hybrid Batch-Quantized Kalman filter (H-BQKF) and the Hybrid Iteratively-Quantized Kalman filter (H-IQKF), that enable the HEF to process local analog measurements and remote measurements quantized to any number of bits, resulting in substantial accuracy improvement. We present extensive simulation and experimental results which demonstrate that the proposed multi-bit hybrid filters outperform the existing single-bit hybrid filter, and obtain performance that is comparable to the standard Kalman filter.