Search for: All records

Abstract—In many scenarios for informative path planning done by ground robots or drones, certain types of information are significantly more valuable than others. For example, in the precision agriculture context, detecting plant disease outbreaks can prevent costly crop losses. Quite often, there is a limit on the exploration budget, which does not allow for a detailed investigation of every location. In this paper, we propose Learned Adaptive Inspection Paths (LAIP), a methodology to learn policies that handle such scenarios by combining uniform sampling with close inspection of areas where high-value information is likely to be found. LAIP combines Q-learning in an offline reinforcement learning setting, careful engineering of the state representation and reward system, and a training regime inspired by the teacher-student curriculum learning model. We found that a policy learned with LAIP outperforms traditional approaches in low-budget scenarios. 

We optimize the overall energy consumption of a Narrowband Internet of Things (NB-IoT) application created using a hybrid blockchain framework. We accomplish this by engineering the underlying hash function (SHA-256) that is used in different procedures (Unique ID generation, Device Join, and Device Transaction) of the blockchain-based NB-IoT system. In order to reduce the complexity of hash verification, IoT devices in the NB-IoT application are built to save the hashes of their authorized transactions as a linear hash chain rather than the entire Merkle tree. Furthermore, base station memory is dynamically partitioned to improve memory usage efficiency and scalability. Compared with the state-of-the-art approach, our approach considerably reduces the total energy consumption of the state-of-the-art application.