The demonstrator focused on indoor asset tracking has been developed as a first functional prototype and successfully tested. The target of the work is focusing on indoor asset localisation based on Bluetooth wireless technology. The focus is on complex indoor environments where Global Navigation Satellite System (GNSS) based solutions are not usable. The idea behind the proposed demonstrator is that the functionality of low-cost and low-energy Bluetooth-based components will be enhanced by AI technology designed to analyse the non-deterministic signal received from tracking tags.
Beyond state-of-the-art developments and impacts
Compared to previous implementations of Bluetooth-based indoor localisation systems developed for testing and evaluation, we consider the AI application the major changing factor regarding system upgradeability and performance improvements. With the introduction of Neural network-based algorithms, the system performance rate also depends on the algorithm structure but mainly on the training process and thus inherently on the quality of training data.