Deep convolutional neural networks (DCNNs) have become one of the most popular approaches to many visual processing tasks. The majority of existing works on the accelerating DCNNs focus on high performance while neglecting the hardware resource utilization, like on-chip memory and DSP. In this paper, we propose a resources-efficient and configurable DCNN accelerator. A fourlevel processing-element (PE)-array-based structure is presented to realize high parallelism calculation of convolutional operation, and a new storage pattern named hybrid stationary (HS) is proposed to take full advantage of the used on-chip memory footprint and limited off-chip memory bandwidth. Moreover, roofline model is adopted to explore the design space of the given hardware resources. The proposed architecture achieves 113 G-ops/s at 100 MHz and consumes 784 DSP48 modules and 211.5 Block RAM modules on ZYNQ-7 ZC706 evaluation board. To the best of our knowledge, the proposed accelerator is the only implemented system on FGPA platform that can achieve multiple advantages: high-performanced, configurable, and efficient in power and resources utilization. It shows significant utilization improvement compared with the other available architectures.

No abstract

No abstract

No abstract

Logistics and transportation industry is not only a major energy consumer, but also a major carbon emitter. Developing green logistics is the only way for the sustainable development of logistics industry. One of the main factors of environmental pollution is caused by carbon emissions in the process of vehicle transportation, and carbon emissions of vehicle transportation is closely related to routing, time-dependent speed and the slope of road. Therefore, vehicle routing problem with time windows considering carbon emissions is presented in this paper. a mixed integer programming model is built to describe the carbon emission optimization problem under the constraint of time windows. In this programming model, the high-granularity predictive speeds are used to compute carbon emissions. And to solve this problem, a hybrid genetic algorithm with adaptive variable neighborhood search method is presented. A case study with the logistics and traffic data in Jingzhou, China is validated, and the results shows that the effectiveness of the proposed model and algorithm.