Analyzing the effectiveness of rescheduling and Flexible Execution methods to address uncertainty in execution duration for a planetary rover
Analyzing the effectiveness of rescheduling and Flexible Execution methods to address uncertainty in execution duration for a planetary rover
复制标题
分析重新安排和灵活执行方法的有效性，以解决行星漫游者执行持续时间的不确定性
DOI：
10.1016/j.robot.2021.103758
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Stephen Kuhn
中科院分区：
文献类型：
--
作者：
Jagriti Agrawal;Wayne Chi;Steve Ankuo Chien;G. Rabideau;D. Gaines;Stephen Kuhn
在执行期间，活动持续时间可能与生成的计划中预测的持续时间不同。在本文中，我们研究（重新）调度调用，在重新调度期间执行，以及灵活的执行，以实现对活动执行持续时间的不确定性的高水平响应。我们讨论这些方法理论上的嵌入式调度程序的上下文中，实际上在有限的CPU嵌入式调度程序与非零调度运行时间的行星rovers.We使用的概念提交窗口，使以前生成的时间表的执行，而（重新）调度的上下文中。我们定义了Fixed Cadence和Event Driven scheduling作为决定何时重新调用调度器的方法。我们定义和分析灵活的执行（FE）作为一种方法来执行生成的时间表，同时使其适应执行的变化。具体来说，FE关注（1）如何利用比预期提前结束的活动，以及（2）如果活动花费的时间比预期多，如何保持一致的时间表。我们提出了一个理论模型和实证结果，记录了这些不同的方法如何相互作用，并对NASA的下一个行星探测器，火星2020探测器的合成数据和最佳可用数据进行处理。然后，我们描述了这些分析如何影响火星2020漫游者的机载软件。
During execution, activity durations may vary from those predicted in the generated schedule. In this article we study (re) scheduling invocation, execution during rescheduling, and flexible execution to enable a high level of responsiveness to uncertainty in activity execution duration. We discuss these methods theoretically in the context of an embedded scheduler and practically in the context of a limited CPU embedded scheduler with a nonzero scheduler runtime intended for a planetary rover.We use the concept of a commit window to enable execution of the previously generated schedule while (re) scheduling. We defineFixed CadenceandEvent Drivenscheduling as methods to decide when to reinvoke the scheduler. We define and analyzeFlexible Execution (FE)as an approach to execute the generated schedule while adapting it to variations in execution. Specifically, FE focuses on (1) how to take advantage of activities ending earlier than expected and (2) how to maintain a consistent schedule if activities take more time than expected. We present a theoretical model and empirical results documenting how these various methods interact and perform on both synthetic data and best available data for NASA’s next planetary rover, the Mars 2020 rover. We then describe how these analyses influenced the onboard software for the Mars 2020 rover.