Ensembles of Self-Aware Robots

The swarm-robotics case study considers ensembles of cooperating, self-aware robots. Robot swarms generally consist of relatively cheap robots that lack many of the capabilities of larger, more sophisticated models, but that can collaborate to collectively achieve tasks that no individual robot in the swarm could accomplish. Robot swarms are particularly well suited for operations in difficult environments where the risk of failure for individual robots is high. For example, robot swarms might be used to rescue victims of natural disasters or industrial accidents. In these situations, the attrition rate of individual robots may be high, but a swarm of robot can continue to function even when individual robots have failed.

In ASCENS we will mainly investigate scenarios where the goal of the robot swarm is to localize and transport objects; this task is similar to the one presented in the initial scenario and also to the rescue task. The objects will be defined in such a way that their transport requires physical cooperation of several robots. This task will have to be performed in a number of increasingly challenging environments containing obstacles such as holes, hills, barriers, bridges, slopes, etc.

Resource Ensembles as Science Clouds

The Science Cloud case study is about making cloud computing more dynamic and open while attempting to maintain its properties of being a reliable and flexible approach for using third-party resources and services, something that is done by both companies and private users using commercial and in-house clouds.

Our goal is to create a platform-as-a-service (PaaS) solution for data sharing and execution of distributed applications, which is based on autonomous, cooperating computers that provide their storage and computational resources on a best-effort and at-will basis. In particular, computers contributing to this platform are not required to stay available for any pre-determined amount of time and may join or leave the system at any given moment. We will explore this property in the form of ad-hoc cooperations between researchers from different organizations. Mechanisms to deal with transiently available computers are a requirement for a usable platform and will therefore be in the focus of the case study, resulting in a highly fault-tolerant solution that will also yield better services compared to those delivered from traditional environments, such as data centers which typically have to deal with failing servers.

Ensembles of Cooperative E-Vehicles

The e-mobility case study aims at illustrating the theories and methodologies developed in ASCENS in the domain of e-mobility planning. Driver, vehicle and infrastructure are considered as interacting autonomous Service Components, which are temporally re/organized in Service Component Ensembles to reach a goal. In order to provide a user with a seamless daily travel plan, a sequence of destinations with possibly different travel modes and resource requirements have to be scheduled. The main intention is to provide benefits for the individual vehicle and its driver as well as for the whole ensemble of vehicles.

Particular attention is paid to the performance criterion of providing a high-level Quality of Service (QoS) that incorporates the following features: Reliability (e.g. transport/delivery reliability, adherence to schedules, guarantee to reach the goal, recharging-in-time assurance), adaptation to changes (e.g. traffic flow, daily personal schedule of the driver) and predictability (confidence in reaching a desired location at a requested time).