We propose EMS (R), a cloud-enabled massive computational experiment management system supporting high-throughput computational robotics research. Compared to existing systems, EMS (R) features a sky-based pipeline or...
详细信息
ISBN:
(纸本)9798350323658
We propose EMS (R), a cloud-enabled massive computational experiment management system supporting high-throughput computational robotics research. Compared to existing systems, EMS (R) features a sky-based pipeline orchestrator which allows us to exploit heterogeneous computing environments painlessly (e.g., on-premise clusters, public clouds, edge devices) to optimally deploy large-scale computational jobs (e.g., with more than millions of computational hours) in an integrated fashion. Cornerstoned on this sky-based pipeline orchestrator, this paper introduces three abstraction layers of the EMS (R) software architecture: (i) Configuration management layer focusing on automatically enumerating experimental configurations;(ii) Dependency management layer focusing on managing the complex task dependencies within each experimental configuration;(iii) Computation management layer focusing on optimally executing the computational tasks using the given computing resource. Such an architectural design greatly increases the scalability and reproducibility of data-driven robotics research leading to much-improved productivity. To demonstrate this point, we compare EMS (R) with more traditional approaches on an offline reinforcement learning problem for training mobile robots. Our results show that EMS (R) outperforms more traditional approaches in two magnitudes of orders (in terms of experimental high throughput and cost) with only several lines of code change. We also exploit EMS (R) to develop mobile robot, robot arm, and bipedal applications, demonstrating its applicability to numerous robot applications.
This study examines the impact of grouping by gender and group roles on robotics performance, computational thinking skills, and learning motivation towards computer programming. One hundred ninety-one students in fou...
详细信息
This study examines the impact of grouping by gender and group roles on robotics performance, computational thinking skills, and learning motivation towards computer programming. One hundred ninety-one students in fourth and fifth grade completed the project. LEGO Mindstorms EV3 robotics were used to compose and program music in groups of three to four students. The robotics project was completed over the course of fourteen weeks for 1 h each week. Gender-structured groups of all female, all male and mixed female and male were randomly assigned in each class. All groups in participating six classrooms were assigned one of three group roles implementation, which were fixed, rotating and no role. This study is significant towards identifying group scaffolds and supports that can produce benefits for all students in robotic activities. Results indicated that group roles matter when students are working on collaborative robotics projects. Three different implementations of group roles in robotics activities demonstrated an impact on robotics performance scores. Gender composition of the group did not impact groups' robotics performance. Group roles also impacted student computational thinking skills, while gender composition of the group still demonstrated no significant difference. Finally, while group roles demonstrated a difference in learning motivation towards computer programming, gender composition demonstrated no difference in learning motivation towards computer programming.
We present an algorithm for determining the shortest path between any two points along the surface of a polyhedron which need not be convex. This algorithm also computes for any source point on the surface of a polyhe...
详细信息
We present an algorithm for determining the shortest path between any two points along the surface of a polyhedron which need not be convex. This algorithm also computes for any source point on the surface of a polyhedron the inward layout and the subdivision of the polyhedron which can be used for processing queries of shortest paths between the source point and any destination point. Our algorithm uses a new approach which deviates from the conventional ''continuous Dijkstra'' technique. Our algorithm has time complexity O(n(2)) and space complexity Theta(n).
暂无评论