Abstract: This paper describes the Chimera II Real-time Operating System, which has been developed for advanced sensor-based control applications. It has been designed as a local operating system, to be used in conjunction with a global operating system. It executes on one or more single board computers in a VMEbus-based system. Advanced sensor-based control systems are both statically and dynamically reconfigurable. As a result, they require many special features, which are currently not found in commercial real-time operating systems. In this paper, we present several design issues for such systems, and we also present the features we have developed and implemented as part of Chimera II. These features include a real-time kernel with dynamic scheduling, global error handling, user signals, and two levels of device drivers; an enhanced collection of interprocessor communication mechanisms, including global shared memory, spin-locks, remote semaphores, priority message passing, global state variable tables, multiprocessor servo task control, and host workstation integration; and several support utilities, including a UNIX C and math libraries, a matrix library, a command interpreter library, and a configuration file library. Chimera II is currently being used with a variety of systems, including the CMU Direct Drive Arm II, the CMU Reconfigurable Modular Manipulator System, the Troikabot System for Rapid Assembly, and the Self-Mobile Space Manipulator. I.

Current systems are very often based on largescale, unpredictable and unreliable infrastructures. However, users of these systems increasingly require services with timeliness properties. This creates a di#cult-to-solve contradiction with regard to the adequate time model: synchronous, or asynchronous? In this paper, we propose an architectural construct and programming model, which address this problem. We assume the existence of a component that is capable of executing timely functions, however asynchronous the rest of the system may be. We call this component the Timely Computing Base, and it can be used by the other components to execute a set of simple but crucial time-related services. We also show how to use it to build dependable and timely applications exhibiting varying degrees of timeliness assurance, under several synchrony models.

requirements, which in essence calls for synchronous system models. However, systems often rely on unpredictable and unreliable infrastructures, that suggest the use of asynchronous models. Several models have been proposed to address this issue. We propose an architectural construct that takes a generic approach to the problem of programming in the presence of uncertain timeliness. We assume the existence of a component, capable of executing timely functions, which helps applications with varying degrees of synchrony to behave reliably despite the occurrence of timing failures. We call this component the Timely Computing Base, TCB. This paper describes the TCB architecture and model, and discusses the application programming interface for accessing the TCB services. The implementation of the TCB services uses fail-awareness techniques to increase the coverage of TCB properties.

Distributed fault-tolerant real-time system models have exhibited a trend to polarize themselves in extreme positions. In this paper, we assess the fitness of current models to represent the attributes underlying the distributed fault-tolerance of real-time systems. Namely, we are concerned with the correctness issues arising from the temporal properties of interprocess communication: reliable and ordered group communication, replication management protocols, time services, etc. We are particularly concerned with best-effort or mission-critical systems, where despite the hard need to fulfil timing guarantees, this cannot be ensured at all times in a given operational envelope, mandating that the system be highly dynamic and adaptive. In the paper, we suggest quasi-synchronism as a framework to address this kind of systems. We finalize by pointing out some contributions to materialize the model. 1 Introduction Fault-tolerant real-time system models, or in the context of this paper, ...

Real-time behavior is materialized by timeliness specifications, which in essence call for synchronous system models. However, systems many often rely on largescale, unpredictable and unreliable infrastructures, that suggest the use of asynchronous models. Several models in between have addressed these antagonistic aims, each in its own way. We propose an architectural construct that addresses the problem in a generic way. We assume the existence of a component that is capable of executing timely functions, however asynchronous the rest of the system may be. This component can be used by other components to execute timely services. There is a certain analogy to the trusted computing base principle used in security. We call it the Timely Computing Base, TCB. In this paper, we show that a TCB can be used to build dependable and timely applications exhibiting varying degrees of timing fault tolerance, under several synchrony models.

The paper focuses on method invocation of real-time objects in a CAN-based distributed real-time system. A simple object model is introduced, which allows the convenient modelling of hardware and software components. Related to the object model, two issues are discussed. Firstly, a model is introduced which allows to form and address object groups. This reflects a basic need in a real-time system to distribute information to multiple clients efficiently. Secondly, we discuss an approach to express timing requirements for object invocations. To achieve distributed consensus on communication resource access, an EDF-like approach is introduced, which takes advantage of knowledge about deadlines, the number of remaining communication activities, and the remaining worst-case execution time for the invoked method at each point of time.

This paper describes current research in real time operating systems. Due to its importance to real-time systems, we begin this survey with a brief summary of relevant results in realtime scheduling and synchronization. Real-time operating systems are described in terms of the primitives and constructs offered to application programs. In addition, the effects of underlying computer architectures on real-time operating systems are discussed, followed by a description of benchmarks and evaluation methods for real-time systems. College of Computing Georgia Institute of Technology Atlanta, Georgia 30332--0280 Readers' comments and suggestions for improvement are solicited. Please direct them to kaushik@cc.gatech.edu. 1 Functionality and Characteristics of Real-time Systems The embedded computer hardware of modern robots and industrial control systems is becoming increasingly complex. Typically, it consists of many interconnected computers operating at multiple levels of control or sup...

This paper addresses the problem of achieving predictable real-time behavior of distributed databases, over environments that do not exhibit hard real-time properties. Unlike soft real-time systems, our quasisynchronous approach aims at achieving hard real-time properties, while given assumptions about the environment hold. As failures get more severe, applications pull back to successively more degraded operational envelopes, that still allow timing assumptions to hold. Timing failure detection and adaptive group communication protocols make this approach effective. The approach is specially useful for telecommunication information network databases, which have to support emerging services that are ever demanding in terms of availability, performance and responsivity. We argue that a soft real-time approach does not deliver the quality of service expected of these databases. In this paper, we use this example application to illustrate how our approach and architecture can support thes...

We are now at the point where the emergence of a new class of applications that operate independently of direct human control can be envisaged However, this is also the crossroads between the requirements put on system support, by the advances of research on high-level models for this class of applications--- e.g. on autonomous agents and distributed AI--- and the shortcomings of current architectures and middleware models.

The paper describes a real-time event channel model in a publisher/subscriber communication scheme. The model specifically considers temporal and reliability attributes and suggests an API that integrates the real-time aspects in the event channel model. According to the need in most real-time systems, we support event channels with different timeliness and reliability classes. Hard real-time event channels are considered to meet all temporal requirements under the specified fault assumptions. The resource requirements for this type of channel are statically assigned by an appropriate reservation scheme. Soft real-time event channels are scheduled by their deadlines, but they are not guaranteed under transient overload conditions. Non real-time event channels are used for events without any specified timeliness requirements in a best-effort manner. The paper finally presents how the different channel classes are mapped to the mechanisms necessary to implement the model on the CAN-Bus.