This paper proposes a pose estimation approach of an industrial six-degree-of-freedom robot without the need of externally placed sensors. An RGB-D camera that is placed at the robots end-effector is combined with a SLAM algorithm to act as a sensor system. The presented system based on the novel integration approach is evaluated by performing test trajectories and is capable of estimating the tool-center point with a standard deviation of 2.6 mm and performing a joint state estimation with a standard deviation of 13.6 mrad without any external sensor.

Depth-data-based sensor systems, such as depth cameras or LiDAR systems, are gaining popularity in the field of robotics, especially in human-robot collaboration. To avoid collisions with humans or external objects, object detection and tracking in the workspace is needed. This paper presents an integrated object cluster tracking and velocity estimation method that is purely based on depth data. Therefore, a tracking heuristic based on similarity and the velocity of the object is used to enable the tracking of external objects. To obtain the velocity, a Kalman filter utilizing a constant velocity model is implemented. For experimental verification, a case study comprising two objects moving within the robot workspace is designed. The experimental setup allows for the initial tracking a maximal trackable object velocity of 9 m s −1, and for already tracked objects a velocity deviation of 3.4 m s-1 to correctly track both repetitive and arbitrary motions of the test objects, and thus constitutes the proposed integrated object cluster tracking approach as a foundation for collision avoidance strategies in robotic tasks.

The measurement of external forces in Collaborative Robots (Cobots) is commonly realized by dedicated sensors in the joints. Since the use of these sensors is cost-intensive and reduces the stiffness of the manipulator, alternative approaches are needed. This paper presents a sensing principle for measuring external forces acting on a robot arm using a generalized momentum observer. The proposed approach does not require dedicated force/torque sensors and relies on measurements available in standard robot joints. The presented method estimates the external forces based on motor current measurement and joint torsion measurement. Both measurements are implemented and compared against a dynamometer in terms of accuracy. Depending on the joint, an RMS error within 2% to 4% of the dynamometer measurement can be achieved.

IEC 61499 is currently explored at a rapid pace, although only few studies were conducted on the role of FPGAs for automation systems based on IEC 61499. With the necessary time quantization for synchronous FPGA systems, a new problem arises – the phenomenon of simultaneous events. Additionally, fan-ins of events complicate the requirements for the system, as it must be ensured that all arriving events are considered. This work proposes an implementation for concurrent events for FPGAs in the context of IEC 61499. To overcome the described problem, the incoming events are split into several single bit signals and are stored in a ring buffer. The proposed approach is verified in simulations of typical situations, testing the behavior of the system with respect to simultaneous events.

Today's markets are short-lived and volatile. The producing industry needs to be able to react fast and cost-effective to product and process changes. Following the trend of Industry 4.0, current automation systems are designed as Cyber-Physical Production System (CPPS), which main properties are the distribution of capabilities and strict modularization to enhance reuse of off-the-shelf automation components. The currently dominant Programmable Logic Controller programming standard IEC 61131 was conceived with mainly centralistic systems in mind, which limits its use for nowadays CPPS scenarios. Focusing on the software aspect, the contribution of this work is an enhanced hierarchic development approach for automation systems, which aims to decouple all hardware from the control flow logic of an automation system for combined IEC 61131 and IEC 61499 systems. Therefore, a migration path for IEC 61131 Function Blocks into the development model of the event-driven IEC 61499 is presented, and issues are discussed. Both approaches pay special attention to the reusability of software.

Automation system engineering becomes more complex, due to the trend toward more flexible, reconfigurable, and modular design approaches, like Industry 4.0. For the modeling and design of the according software, two standards are present: IEC 61131-3 and IEC 61499. In order to satisfy the requirements for modern, large scale, highly distributed applications while also supporting still existing legacy systems, the demand for a combined development framework arises, where the best of breed tool can be chosen for a given automation task. Considering that, the IEC 61499 model is extended to allow the dual development and execution of IEC 61131-3 programs, and enabling easy and correct interaction between the two paradigms. In order to verify the validity of the chosen approach, an IEC 61499 development tool and a runtime environment is modified to support IEC 61131-3. A sample application is implemented, which comprises a pure IEC 61131-3 part with a 1 ms cycle time, a pure IEC 61499 part, and a part with interaction between both subparts, in order to evaluate possible interference between the runtime parts. Experimental results show that no interference is occurring, and the chosen development approach allows the seamless integration of IEC 61131-3 and IEC 61499 in one combined development framework.

The IEC 61131-3 and IEC 61499 standards are currently used to implement automation systems. Although aspects as reuseability, execution models, and component-based design have been discussed and analyzed for both standards, the programming effort and complexity of implementation for typical tasks in industrial automation have not been compared so far. This work proposes two typical application classes of industrial automation systems and implements one concrete application for each class with IEC 61131-3 and IEC 61499, respectively. Both implementations are then evaluated via McCabe's Cyclomatic Complexity and Halstead software measures, providing an objective comparison on the implementations.

In this paper, a framework for the automatic generation of redundancy-free interlock code is proposed. The core component is represented by an ontology-based integration framework that holds the relevant information from a developed piping and instrumentation diagram (P&ID) editor as well as the PLC program where the interlocks should be added. The integration framework is mainly based on AutomationML concepts as glue model between the different internal models of the P&ID editor and the PLC programming tool. In order to enable the coexistence of manual and automatically generated interlocking code, the approach is extended by additionally applying a theorem prover. Here, the theorem prover Z3 is utilized to eliminate redundant interlock logic whereas the redundancy-free logic is then automatically added as interlock code to the IEC 61131-3 compliant program. The framework is evaluated by application to a laboratory process plant.

In this paper, a framework for automatic generation of a flexible and modular system for fault detection and diagnosis (FDD) is proposed. The method is based on an ontology-based integration framework, which gathers the information from various engineering artifacts. Based on the ontologies, FDD functions are generated based on structural and procedural generation rules. The rules are encoded as SPARQL queries which automatically build logical segments of the entire manufacturing system in the ontology, assign sub-processes to these segments, and finally generate the appropriate FDD system for the sub-process. These generated modular FDD functions are additionally combined in a modular way to enable the fault detection and diagnosis of the entire system. The effectiveness of the approach is demonstrated by a first application to a conveyor system.

The demand for flexible automation architectures in industry is raising, due to the trend to individualized products and the shortening of product life-cycles in general. Current approaches in automation systems are either only suited for a small range of processes or are only reprogrammable by software and automation experts. The presented domain specific language based approach decouples production-domain concerns from the automation program, by separating the process description into a specialized domain language. This domain language can be used by plant owners and domain experts to define and execute vastly different processes, as long as the needed process steps are supported by the actual automation system. This contribution demonstrates the development of a domain-centric SOA-based batch process automation system, by creating a directly executable DSL based on domain terms found in ISA-88.

In this paper, an ontology-based approach to automatically generate control applications to handle diagnostic information of decentralized control devices is presented. Diagnostic possibilities of modern remote I/O devices are analyzed and software components in terms of function blocks to handle the specific diagnostic information are defined. After a detailed conceptual overview, the application of the proposed knowledge-based code generation approach to a PLC-based control architecture of a hot rolling mill is described. It is shown that the proposed approach significantly reduces engineering time and the error rate in the design processes of industrial control and diagnostic applications, since the application engineering is raised to an abstract level by utilizing pre-defined, tested, and reusable function blocks and a user-definable set of code generation rules to encode repetitive implementation tasks. The rules are defined in the query language SPARQL with additional ARQ functions to reduce the complexity.

Reusability of control code is crucial for future industrial automation in order to allow the development of high quality control systems in less time and with less cost. A key stopper for automation software reuse is the current intertwining of control logic with hardware specific code. Existing approaches that tackled such a problem could lead to improvements but will not sufficiently solve this issue. In this work we investigate a concept based on a logical grouping of I/Os to so called Instrumentation- and Control-Points and decoupling them from the control application by applying IEC 61499's adapter concept. In two industrial motivated application examples we show that this concept fulfills the requirements for increasing the reusability of control code by separating control logic from hardware specific code. Moreover, it leads to a better application software structure which is easier to maintain and to understand.

Automation systems (ASs) based on Service-Oriented Architecture (SOA) are becoming more wide-spread. Due to their inherent property of separating the hardware-specific implementation from the logical work- or control-flow, SOA-based application programming is regarded as the next logical step in AS programming. Several successful attempts for implementing SOA-based AS have been made, but always with non-industrial, non-standard software like Devices Profile for Web Services. Based on the requirements of typical industrial SOA-based AS application, we will show that the industry standard data acquisition software OPC Unified Architecture is capable to serve as a SOA-based middleware.

Lack of awareness about the potentials of agent technology, as well as concerns regarding stability, scalability, and survivability especially in unpredictable environments of attacks and system failures are identified as major reasons for missing trust in the idea of delegating tasks to autonomous agents. Therefore, one of the key research issues is how the quality of Multi-Agent Systems can be assured, while at the same time reducing the development efforts to enhance the acceptance of agent technology in the industry. In order to ensure the industrial relevance, we focus on development of methods and techniques for the evaluation and tests of agent-based control applications. Our research covers the development of demonstration framework, including pilot implementations in industriallike settings.

Modern industrial automation systems engineering (ASE) environments have to accommodate for heterogeneity coming from the engineering disciplines involved, the software tools and their data models, and run-time data collection. In many ASE environments domain experts have to invest considerable effort to bridge the semantic gaps between common project-level engineering concepts and the diverse local data representations. In this paper we discuss the needs for semantic integration and applications of machine-understandable knowledge engineering in three real-world ASE use cases from our industry partners. We provide an evaluation concept with empirical studies to measure the benefits and limitations of the proposed approach compared to the traditional expert-intensive approach. Major result of the initial evaluation is that semantic integration has good potential to make engineering processes more efficient and robust if supported well with user interfaces that end users find usable and useful.

Current component-based approaches for distributed control systems enable the reuse of mechatronic components, for example specific grippers or conveyor belts, but do not allow full application reuse as the actual hardware configuration is an implicit part of the software solution. In contrast to the current object-oriented or component-based engineering models for automation engineering, this paper will introduce a new Model-Driven Architecture based application and hardware modeling approach. Through separation of the logical application domain and the specific hardware domain the logical application is completely independent of the concrete physical plant configuration. Therefore development time for automation applications can be vastly reduced and software quality can be improved.

Traditional fault detection systems in industrial control applications are just able to report occurring faults. Fault diagnosis systems are more desirable for plant operators, as such systems are capable to reduce the number of occurring alarms by elimination of consecutive alarms and prioritization of critical alarms. The disadvantage of those systems is that they have to be implemented anew for every control application, as the system dependencies vary from application to application. Ontology-based fault diagnosis systems do not have this disadvantage. Only the ontology has to be created for the new system, which greatly reduces time and effort for new systems, as old ontologies can be reused for the new system.