A2E2: Aerial-assisted energy-efficient edge sensing in intelligent public transportation systems

Emerging intelligent public transportation systems (IPTS) enable various smart services by collecting sensing data leveraging public transportation. However, existing researches ignore sensors are usually deployed in 3-D space, and a number of generated sensing data are dropped due to limited storage capacities of sensors. To solve the problem, we propose the A2E2 framework to ensure the sustainable sensing operation, i.e., collecting data and charging sensors assisted by UAVs in IPTS simultaneously. In A2E2 framework, UAVs could charge sensors and fly near them to collect sensing data. Meanwhile, they also could stop on vehicles to be charged and send collected data to smart services. The entire process is formulated as a collaborative task allocation and scheduling jointly optimization problem. We solve the problem by dividing it into two parts which are proved NP-hard, and propose two algorithms, i.e., the minimum transfer algorithm and shortest trajectory planning algorithm. The minimum transfer algorithm minimizes the number of UAV transfer times among vehicles, and the shortest trajectory planning algorithm ensures UAVs could execute the allocated task effectively. Extensive evaluations with large scale real world dataset are conducted showing that A2E2 outperforms the other three benchmarks significantly.