No abstract

In building intelligent transportation systems such as taxi or rideshare services, accurate prediction of travel time and distance is crucial for customer experience and resource management. Using the NYC taxi dataset, which contains taxi trips data collected from GPS-enabled taxis [1], this paper investigates the use of deep neural networks to jointly predict taxi trip time and distance. We propose a model, called ST-NN (Spatio-Temporal Neural Network), which first predicts the travel distance between an origin and a destination GPS coordinate, then combines this prediction with the time of day to predict the travel time. The beauty of ST-NN is that it uses only the raw trips data without requiring further feature engineering and provides a joint estimate of travel time and distance. We compare the performance of ST-NN to that of state-of-the-art travel time estimation methods, and we observe that the proposed approach generalizes better than state-of-the-art methods. We show that ST-NN approach significantly reduces the mean absolute error for both predicted travel time and distance, about 17% for travel time prediction. We also observe that the proposed approach is more robust to outliers present in the dataset by testing the performance of ST-NN on the datasets with and without outliers.

This paper considers the problem of generating the shortest time division multiple access (TDMA) schedule for use in rechargeable wireless sensor networks (rWSNs) with heterogeneous energy arrivals rates. This novel problem considers: 1) the time required by nodes to harvest sufficient energy to transmit/receive a packet; 2) harvest-use-store (HUS) energy harvesting and usage models, and; 3) battery imperfections, i.e., leakage, storage efficiency, and capacity. This paper shows the problem at hand, called link scheduling in harvest-use-store (LSHUS), is in general NP-Complete. Furthermore, it presents a greedy heuristic, called LS-rWSN, to solve LSHUS. Our experiments show that a longer energy harvesting time (leakage rate) from 1 to 20 (0% to 4%) increases the schedule length by up to 565.82 (44.54%) slots while reducing storage efficiency from 1.0 to 0.6 lengthens the schedule by up to 62.77%. In contrast, battery capacity has an insignificant effect, i.e., enlarging the capacity by 20 times decreases the schedule length by only 6.5%.

No abstract