| 1. |
- Jorge, Emilio, 1992, et al.
(author)
-
Reinforcement learning in real-time geometry assurance
- 2018
-
In: Procedia CIRP. - : Elsevier BV. - 2212-8271. ; 72, s. 1073-1078
-
Conference paper (peer-reviewed)abstract
- To improve the assembly quality during production, expert systems are often used. These experts typically use a system model as a basis for identifying improvements. However, since a model uses approximate dynamics or imperfect parameters, the expert advice is bound to be biased. This paper presents a reinforcement learning agent that can identify and limit systematic errors of an expert systems used for geometry assurance. By observing the resulting assembly quality over time, and understanding how different decisions affect the quality, the agent learns when and how to override the biased advice from the expert software.
|
|
| 2. |
- Riazi, Sarmad, 1986, et al.
(author)
-
Parallelization of a gossip algorithm for vehicle routing problems
- 2018
-
In: IEEE International Conference on Automation Science and Engineering. - 2161-8070 .- 2161-8089. ; 2018-August, s. 92-97
-
Conference paper (peer-reviewed)abstract
- A large body of research on the vehicle routing problem and its variants focuses on developing efficient solution procedures. Yet, not so many research articles have addressed parallelism in their proposed algorithms. Parallelized optimization algorithms can yield better solution quality in less amount of time. The main contribution of this paper is parallelization of a distributed algorithm based on the gossip protocol for vehicle routing problems VRP. The proposed algorithm can be applied to different variants of VRPs. While the resulting speed-ups are promising, the required effort for implementation of the parallelism is minimal, which makes the algorithm even more appealing.
|
|
| 3. |
- Lennartson, Bengt, 1956
(author)
-
Reductions and Abstractions for Optimization of Modular Timed Automata
- 2022
-
In: IFAC-PapersOnLine. - : Elsevier BV. - 2405-8963 .- 2405-8963. ; 55:28, s. 344-349
-
Conference paper (peer-reviewed)abstract
- Time optimization of concurrent sequences of operations is in this paper solved by timed automata. To reduce the complexity of this classical problem, including applications such as planning and scheduling, an abstraction method has recently been proposed based on local optimization (Hagebring and Lennartson, 2019). In a modular subsystem, local paths without any communication with other subsystems are optimized with respect to time, and when subsystems are synchronized more local behavior appears. The proposed method has shown to be successful, drastically reducing computational complexity for important classes of planning problems. The only drawback is that the synchronous composition includes a heuristic non-standard synchronous composition procedure to achieve true con currency. In this paper a simple solution to this problem is presented based on the original synchronous composition of timed automata. In the transformation of the timed automaton to an ordinary automaton, where time weights are generated, it is first observed that the state space often increases dramatically in this transformation. To solve this complexity problem, an efficient reduction is proposed as a complement to local optimization, and both methods are demonstrated to be very efficient when they are applied to realistic benchmark examples. Copyright (C) 2022 The Authors.
|
|
| 4. |
- Bengtsson, Kristofer, 1979, et al.
(author)
-
Aspect-Oriented Programming for Manufacturing Automation Control Systems
- 2009
-
In: 13th IFAC Symposium on Information Control Problems in Manufacturing Technology, INCOM’09, Moscow, Russia. - 1474-6670. - 9783902661432 ; :PART 1, s. 668-673
-
Conference paper (peer-reviewed)abstract
- The languages defined by the standard IEC 6-1131 for Programm\-able Logic Controllers (PLCs), have both benefits and drawbacks compared to object-oriented programming languages. For example, the current programming approach is good for maintaining the system, since it is easy to program and comprehend for a plant floor personal with limited programming experience. The main drawbacks are reusability, code dependability and flexibility. This paper proposes an aspect-oriented programming (AOP) approach as an alternative solution, which can keep the benefits and eliminate the drawbacks. AOP is an emerging programming methodology in computer science, which helps the programmer by separating different crosscutting concerns in the program. These crosscutting concerns and tangled behavior are especially important issues in PLC programming, which makes AOP a suitable approach. The paper describes how to adapt the AOP methodology into PLC program development and how it can benefit the control design process.
|
|
| 5. |
- Fei, Zhennan, 1984, et al.
(author)
-
Efficient Symbolic Supervisory Synthesis and Guard Generation: Evaluating partitioning techniques for the state-space exploration
- 2011
-
In: ICAART 2011 - Proceedings of the 3rd International Conference on Agents and Artificial Intelligence. - 9789898425409 ; 1, s. 106-115
-
Conference paper (peer-reviewed)abstract
- The supervisory control theory (SCT) is a model-based framework, which automatically synthesizes a supervisor that restricts a plant to be controlled based on specifications to be fulfilled. Two main problems, typically encountered in industrial applications, prevent SCT from having a major breakthrough. First, the supervisor which is synthesized automatically from the given plant and specification models might be incomprehensible to the users. To tackle this problem, an approach was recently presented to extract compact propositional formulae (guards) from the supervisor, represented symbolically by binary decision diagrams (BDD). These guards are then attached to the original models, which results in a modular and comprehensible representation of the supervisor. However, this approach, which computes the supervisor symbolically in the conjunctive way, might lead to another problem: the state-space explosion, because of the large number of intermediate BDD nodes during computation. To alleviate this problem, we introduce in this paper an alternative approach that is based on the disjunctive partitioning technique, including a set of selection heuristics. Then this approach is adapted to the guard generation procedure. Finally, the efficiency of the presented approach is demonstrated on a set of benchmark examples.
|
|
| 6. |
- Cronrath, Constantin, 1990, et al.
(author)
-
Enhancing digital twins through reinforcement learning
- 2019
-
In: IEEE International Conference on Automation Science and Engineering. - 2161-8070 .- 2161-8089. ; 2019-August, s. 293-298
-
Conference paper (peer-reviewed)abstract
- Digital Twins are core enablers of smart and autonomous manufacturing systems. Although they strive to represent their physical counterpart as accurately as possible, slight model or data errors will remain. We present an algorithm to compensate for those residual errors through Reinforcement Learning (RL) and data fed back from the manufacturing system. When learning, the Digital Twin acts as teacher and safety policy to ensure minimal performance. We test the algorithm in a sheet metal assembly context, in which locators of the fixture are optimally adjusted for individual assemblies. Our results show a fast adaption and improved performance of the autonomous system.
|
|
| 7. |
- Svedlund, Ludvig, 1993, et al.
(author)
-
Co-simulation of Rigid Interactions Using Differential Algebraic Equations
- 2021
-
In: IEEE International Conference on Emerging Technologies and Factory Automation, ETFA. - 1946-0759 .- 1946-0740. ; 2021-September
-
Conference paper (peer-reviewed)abstract
- In order for virtual commissioning to be applied in an industrial setting, it is required that the models used are predefined and standardized. These models also need to be able to protect intellectual property, so that systems containing such information also can be modeled and used. This paper presents and evaluates an alternative method for co-simulation, based on a differential algebraic equation (DAE) formulation. The method specifies how models should be created and used, such that sharing model information between interacting components can be avoided. The focus is specifically on co-simulation of rigidly interacting models, such as robots interacting with other mechanical components.
|
|
| 8. |
|
|
| 9. |
|
|
| 10. |
- Adlemo, Anders, 1957, et al.
(author)
-
Towards a True Flexible Manufacturing System
- 1994
-
In: Proc of the 10th ISPE/IFAC Int. Conference on CAD/CAM, Robotics and Factories of the Future, CARS & FOF '94, Ottawa, Canada. ; , s. 401-410
-
Conference paper (peer-reviewed)
|
|
