Today's large-scale parallel workflows are often processed on heterogeneous distributed computing platforms. From an economic perspective, computing resource providers should minimize the cost while offering high service quality. It has become well-recognized that energy consumption accounts for a large part of a computing system's total cost, and timeliness and reliability are two important service indicators. This work studies the problem of scheduling a parallel workflow that minimizes the system energy consumption under the constraints of response time and reliability. We first mathematically formulate this problem as a Non-linear Mixed Integer Programming problem. Since this problem is hard to solve directly, we present some highly-efficient heuristic solutions. Specifically, we first develop an algorithm that minimizes the schedule length while meeting reliability requirement, on top of which we propose a processor-merging algorithm and a slack time reclamation algorithm using a dynamic voltage frequency scaling (DVFS) technique to reduce energy consumption. The processor-merging algorithm tries to turn off some energy-inefficient processors such that energy consumption can be minimized. The DVFS technique is applied to scale down the processor frequency at both processor and task levels to reduce energy consumption. Experimental results on two real-life workflows and extensive synthetic parallel workflows demonstrate their effectiveness.

Achieving the high accuracy of blind roads and crosswalks recognition is important for blind guiding equipment to help blind people sense the surrounding environment. A lightweight semantic segmentation network is proposed to quickly and accurately segment blind roads and crosswalks in a complex road environment. Specifically, a lightweight network with depthwise separable convolution as a component is used as a basic module to reduce the number of parameters of the model and increase the speed of semantic segmentation. In order to ensure the segmentation accuracy of the network, we use a densely connected atrous spatial pyramid pooling module to extract feature information of different angles and context feature modules to enhance the effectiveness of different levels of feature information fusion. To verify the effectiveness of the proposed method, we collect and produce a data set from a real environment, which contains two objects of blind roads and crosswalks¹. Experimental results demonstrate that, compared to some state-of-the-art approaches, the proposed approach greatly improves the segmentation speed, while achieving better or similar accuracy, which shows that the proposed approach provides a better basis for the application of devices for guiding the blind.

This paper extends a rapidly-exploring random tree (RRT) algorithm to plan the motion for a wheeled robot under kinodynamic constraints. It consists of an RRT extension algorithm that can quickly find an obstacle-free path for a robot to reach a goal, short-cutting algorithm that removes redundant waypoints from a previous obstacle-free path, and collision-based path deformation algorithm that integrates a motion-control law into a path deformation strategy. Unlike previous algorithms that apply complex steer functions during a path sampling phase, we use a straight line to connect a pair of sampled waypoints such that an obstacle-free path can be quickly found. This path is pruned by the short-cutting algorithm. With the pruned path, we propose a motion-control law that is guided by a pose-based steer function for the robot to reach its destination in a short time. A path deformation strategy is presented that shifts the waypoint away from collision point such that the trajectory can be generated without any collision. Results demonstrate that the proposed framework needs less computation time to generate a smoother trajectory with shorter length than its peers, and that simulated trajectories of our controller are close to real ones and outperforms a pose-based controller.

Cloud computing has become an important paradigm in which scalable resources can be provided to users to remotely process their applications. In a cloud computing platform, energy consumption accounts for a significant cost portion. This work thus aims to present an energy-efficient scheduling algorithm for processing a user application with a real-time requirement. This problem is formulated as a non-linear mixed integer programming problem. We start with providing an optimal closed-form solution to its relaxation problem that aims to minimize the energy consumption without considering real-time requirements. To meet real-time requirements, we propose how to adjust task placement and resource allocation by making a good tradeoff between energy consumption and task execution time. Lastly, we propose to adjust the start time of task execution such that the application's completion time can be further shortened. Experimental results on two real-case benchmarks and extensive synthetic applications demonstrate that our proposed method finds a schedule that generally has 30% and 20% less energy consumption than E-HEFT and genetic algorithm, respectively. Besides, the proposed method has a higher rate to successfully find a feasible schedule than them, and its computation time is close to E-HEFT's, but far less than the genetic algorithm.