Robust Scheduling in Team-Robotics

9 Abstract 10 Mobile robots interact with a dynamically changing, physical environment. All tasks controlling such interactions must be per-11 formed reliably and in real-time. Information from the local sensors often is incomplete or inconsistent. Distributed sensor fusion is 12 a technique that enables a team to get a more complete view of the world with a better quality of the provided information. In this 13 paper we address the problem of scheduling the local processing tasks that are part of the overall fusion process. The particular 14 problem to be addressed lies in the unpredictable execution times of these tasks, which do not allow for scheduling using worst-case 15 execution times. The Time-Aware Fault-Tolerant (TAFT) scheduler allows working with expected-case execution times instead, and 16 still achieves a predictable timing behavior. The paper details an efficient scheduling strategy for TAFT based on Earliest Deadline 17 algorithms, formalizing the adopted task model and the underlying scheduling mechanism. Results are presented showing the 18 achieved real-time behavior with an increased acceptance rate, a higher throughput, and a graceful degradation in transient overload 19 situations compared to standard schedulers. Additionally, it describes the implementation of TAFT in the real-time platform that is 20 embedded in our robot team. 25 Team robotics is still a challenge for several disci-26 plines of computer science as well as for electrical and 27 mechanical engineering. This challenge is shown in the 28 RoboCup (Kitano et al., 1997) scenario, where teams 29 of robots play soccer against each other. The design of 30 the mobile, autonomous robots itself is a non-trivial 31 problem. Beyond all electro-mechanical issues it is ongo-32 ing research of computer scientists to develop control 33 algorithms that are powerful enough to navigate and 34 operate a robot in a potentially unknown, often unpre-35 dictable, and dynamic environment. This is even more 36 difficult as due to space and energy constraints the com-37 puting power of mobile, autonomous robots will remain 38 very limited even in the foreseeable future. Additionally, 39 another dimension of complexity enters the scene when 40 these autonomous robots have to cooperate with each 41 other in order to fulfill a given task. Cooperation intro-42 duces the need for communication and synchronization 43 in order to avoid collisions, identify and localize objects 44 etc. As teams of robots are designed to interact with the 45 physical world, all this has …