| 1. |
- Blanch, Krister, 1991
(författare)
-
Beyond-application datasets and automated fair benchmarking
- 2023
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Beyond-application perception datasets are generalised datasets that emphasise the fundamental components of good machine perception data. When analysing the history of perception datatsets, notable trends suggest that design of the dataset typically aligns with an application goal. Instead of focusing on a specific application, beyond-application datasets instead look at capturing high-quality, high-volume data from a highly kinematic environment, for the purpose of aiding algorithm development and testing in general. Algorithm benchmarking is a cornerstone of autonomous systems development, and allows developers to demonstrate their results in a comparative manner. However, most benchmarking systems allow developers to use their own hardware or select favourable data. There is also little focus on run time performance and consistency, with benchmarking systems instead showcasing algorithm accuracy. By combining both beyond-application dataset generation and methods for fair benchmarking, there is also the dilemma of how to provide the dataset to developers for this benchmarking, as the result of a high-volume, high-quality dataset generation is a significant increase in dataset size when compared to traditional perception datasets. This thesis presents the first results of attempting the creation of such a dataset. The dataset was built using a maritime platform, selected due to the highly dynamic environment presented on water. The design and initial testing of this platform is detailed, as well as as methods of sensor validation. Continuing, the thesis then presents a method of fair benchmarking, by utilising remote containerisation in a way that allows developers to present their software to the dataset, instead of having to first locally store a copy. To test this dataset and automatic online benchmarking, a number of reference algorithms were required for initial results. Three algorithms were built, using the data from three different sensors captured on the maritime platform. Each algorithm calculates vessel odometry, and the automatic benchmarking system was utilised to show the accuracy and run-time performance of these algorithms. It was found that the containerised approach alleviated data management concerns, prevented inflated accuracy results, and demonstrated precisely how computationally intensive each algorithm was.
|
|
| 2. |
- Theorin, Alfred, et al.
(författare)
-
An event-driven manufacturing information system architecture for Industry 4.0
- 2017
-
Ingår i: International Journal of Production Research. - : Taylor & Francis. - 0020-7543 .- 1366-588X. ; 55:5, s. 1297-1311
-
Forskningsöversikt (refereegranskat)abstract
- Future manufacturing systems need to be more flexible, to embrace tougher and constantly changing market demands. They need to make better use of plant data, ideally utilising all data from the entire plant. Low-level data should be refined to real-time information for decision-making, to facilitate competitiveness through informed and timely decisions. The Line Information System Architecture (LISA), is presented in this paper. It is an event-driven architecture featuring loose coupling, a prototype-oriented information model and formalised transformation services. LISA is designed to enable flexible factory integration and data utilisation. The focus of LISA is on integration of devices and services on all levels, simplifying hardware changes and integration of new smart services as well as supporting continuous improvements on information visualisation and control. The architecture has been evaluated on both real industrial data and industrial demonstrators and it is also being installed at a large automotive company. This article is an extended and revised version of the paper presented at the 2015 IFAC Symposium on Information Control in Manufacturing (INCOM 2015). The paper has been restructured in regards to the order and title of the chapters, and additional information about the integration between devices and services aspects have been added. The introduction and the general structure of the paper now better highlight the contributions of the paper and the uniqueness of the framework.
|
|
| 3. |
- Xie, Haotian, et al.
(författare)
-
Robust parallel predictive torque control with model reference adaptive estimator for im drives
- 2020
-
Ingår i: Proceedings - 2020 International Conference on Electrical Machines, ICEM 2020. ; 23 August 2020, s. 1219-1224
-
Konferensbidrag (refereegranskat)abstract
- This paper presents the robustness improvement for the proposed parallel structure predictive torque control (PPTC) via a MRA-based estimator. Although predictive torque control (PTC) has the merits of lower switching frequency and straightforward implementation, it inevitably suffers from the inherent drawbacks of high torque ripple and inappropriate tuning of the weighting parameter. To solve this issue, the proposed PPTC employs two homogeneous objective terms which are optimized in a parallel strucutre, to bypass the usage of weighting parameters. However, the parameter mismatches in the control plant will lead to the prediction torque and flux error, which further impacts the control behavior of the system. Therefore, this paper evaluates the parameter sensitivity for PPTC, aiming to improve robustness of the proposed algorithm with a MRA-based parameter estimator. Finally, the validity of the proposed scheme is confirmed through an experimental assessment.
|
|
| 4. |
- Yin, Hang, et al.
(författare)
-
Policy Learning with Embedded Koopman Optimal Control
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- Embedding an optimization process has been explored for imposing efficient and flexible policy structures. Existing work often build upon nonlinear optimization with explicitly unrolling of iteration steps, making policy inference prohibitively expensive for online learning and real-time control. Our approach embeds a linear-quadratic-regulator (LQR) formulation with a Koopman representation, thus exhibiting the tractability from a closed-form solution and richness from a non-convex neural network. We use a few auxiliary objectives and reparameterization to enforce optimality conditions of the policy that can be easily integrated to standard gradient-based learning. Our approach is shown to be effective for learning policies rendering an optimality structure and efficient reinforcement learning, including simulated pendulum control, 2D and 3D walking, and manipulation for both rigid and deformable objects. We also demonstrate real world application in a robot pivoting task.
|
|
| 5. |
- Ge, Yu, 1995, et al.
(författare)
-
5G SLAM Using the Clustering and Assignment Approach with Diffuse Multipath
- 2020
-
Ingår i: Sensors. - : MDPI AG. - 1424-8220. ; 20:16, s. 1-31
-
Tidskriftsartikel (refereegranskat)abstract
- 5G communication systems operating above 24 GHz have promising properties for user localization and environment mapping. Existing studies have either relied on simplified abstract models of the signal propagation and the measurements, or are based on direct positioning approaches, which directly map the received waveform to a position. In this study, we consider an intermediate approach, which consists of four phases-downlink data transmission, multi-dimensional channel estimation, channel parameter clustering, and simultaneous localization and mapping (SLAM) based on a novel likelihood function. This approach can decompose the problem into simpler steps, thus leading to lower complexity. At the same time, by considering an end-to-end processing chain, we are accounting for a wide variety of practical impairments. Simulation results demonstrate the efficacy of the proposed approach.
|
|
| 6. |
- Liu, Yang, et al.
(författare)
-
A Study Towards Reliability- and Delay-Critical Wireless Communication for RoboCup Robotic Soccer Application
- 2007
-
Ingår i: 2007 International Conference on Wireless Communications, Networking and Mobile Computing. - : Institute of Electrical and Electronics Engineers (IEEE). - 1424413117 - 9781424413119 - 1424413125 - 1424413125 - 1424413117 ; , s. 633-636
-
Konferensbidrag (refereegranskat)abstract
- RoboCup is an international competition for a team of multiple fast-moving robots under a dynamic environment to autonomously play soccer game against another team. Due to the nature of the competition, the requirements and constraints for the wireless communication are extremely tight. The challenge is that wireless communication is involved in the control loop and therefore the reliability and propagation delay are vital factors which directly affect the team performance. Beside, various interferences with known and unknown frequency / transmission power usually exist at the competition site, which are hazardous environment to achieve reliable and low latency performance for wireless communication. This paper investigates the performance strengths and weaknesses of wireless communication means e.g. RF, IEEE 802.11a/b, IEEE 802.15.4, DECT, Linx, etc, which are commonly used nowadays in different RoboCup teams and propose improvements towards reliable and delay-critical wireless communication.
|
|
| 7. |
- Ortiz Morales, Daniel, 1984-, et al.
(författare)
-
Path-constrained motion analysis : an algorithm to understand human performance on hydraulic manipulators
- 2015
-
Ingår i: IEEE Transactions on Human-Machine Systems. - 2168-2291. ; 45:2, s. 187-199
-
Tidskriftsartikel (refereegranskat)abstract
- We propose a novel method to analyze how human operators use hydraulic manipulators of heavy-duty equipment. The approach is novel in the sense that it applies knowledge of motion planning and optimization techniques used in robotics. As an example, we consider the case of operating a forestry crane. To that end, we use motion data that has been recorded during standard operation with the help of sensors and a data acquisition unit. The data backs up the notion that operators work by performing repeatable patterns observed in the trajectories of the manipulator's joints. We show how this nominal behavior is computed, and consequently this allow us to present the following: 1) an analytical procedure to analyze motions, 2) how to represent the "performance" of the operator in a 2D plot, 3) an example of how to use this information to suggest customized control settings, and 4) some complementary ideas needed for improving efficiency through automation.
|
|
| 8. |
- Behdadi, Dorna, 1988, et al.
(författare)
-
A Normative Approach to Artificial Moral Agency
- 2020
-
Ingår i: Minds and Machines. - : Springer Science and Business Media LLC. - 0924-6495 .- 1572-8641. ; 30:2
-
Tidskriftsartikel (refereegranskat)abstract
- This paper proposes a methodological redirection of the philosophical debate on artificial moral agency (AMA) in view of increasingly pressing practical needs due to technological development. This “normative approach” suggests abandoning theoretical discussions about what conditions may hold for moral agency and to what extent these may be met by artificial entities such as AI systems and robots. Instead, the debate should focus on how and to what extent such entities should be included in human practices normally assuming moral agency and responsibility of participants. The proposal is backed up by an analysis of the AMA debate, which is found to be overly caught in the opposition between so-called standard and functionalist conceptions of moral agency, conceptually confused and practically inert. Additionally, we outline some main themes of research in need of attention in light of the suggested normative approach to AMA.
|
|
| 9. |
- Krook, Jonas, 1986
(författare)
-
Formal Methods and Safety for Automated Vehicles: Modeling, Abstractions, and Synthesis of Tactical Planners
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- One goal of developing automated road vehicles is to completely free people from driving tasks. Automated vehicles with no human driver must handle all traffic situations that human drivers are expected to handle, possibly more. Though human drivers cause a lot of traffic accidents, they still have a very low accident and failure rate that automated vehicles must match. Tactical planners are responsible for making discrete decisions for the coming seconds or minutes. As with all subsystems in an automated vehicle, these planners need to be supported with a credible and convincing argument of their correctness. The planners interact with other road users in a feedback loop, so their correctness depends on their behavior in relation to other drivers and road users over time. One way to ascertain their correctness is to test the vehicles in real traffic. But to be sufficiently certain that a tactical planner is safe, it has to be tested on 255 million miles with no accidents. Formal methods can, in contrast to testing, mathematically prove that given requirements are fulfilled. Hence, these methods are a promising alternative for making credible arguments for tactical planners’ correctness. The topic of this thesis is the use of formal methods in the automotive industry to design safe tactical planners. What is interesting is both how automotive systems can be modeled in formal frameworks, and how formal methods can be used practically within the automotive development process. The main findings of this thesis are that it is viable to formally express desired properties of tactical planners, and to use formal methods to prove their correctness. However, the difficulty to anticipate and inspect the interaction of several desired properties is found to be an obstacle. Model Checking, Reactive Synthesis, and Supervisory Control Theory have been used in the design and development process of tactical planners, and these methods have their benefits, depending on the application. To be feasible and useful, these methods need to operate on both a high and a low level of abstraction, and this thesis contributes an automatic abstraction method that bridges this divide. It is also found that artifacts from formal methods tools may be used to convincingly argue that a realization of a tactical planner is safe, and that such an argument puts formal requirements on the vehicle’s other subsystems and its surroundings.
|
|
| 10. |
- La Hera, Pedro
(författare)
-
Non-linear dynamics modelling description for simulating the behaviour of forestry cranes
- 2014
-
Ingår i: International Journal of Modelling, Identification and Control. - : Inderscience Publishers. - 1746-6172 .- 1746-6180. ; 21, s. 125 - 138
-
Tidskriftsartikel (refereegranskat)abstract
- Model-based design is a standard framework widely adopted in modern industry. It is used for designing multi-domain engineering solutions based on computer-aided simulation technology. Currently, this approach is also being considered as a tentative method for designing modern heavy-duty machine technology. Under this motivation, our aim is to present how modelling techniques can be used for simulating dynamics of forestry machines. To this end, we consider a forestry crane, and propose mathematical models and calibration techniques, such that model-based methods can subsequently be applied. The complexity of the machine is represented by first principle laws, in which the mechanical system is modelled by Euler-Lagrange formulations, and the hydraulic system is modelled by principles of fluid dynamics. The calibration algorithms are performed by statistical algorithms based on linear and non-linear least-squares methods. The results of simulation show a significant correspondence between the simulated and observed variables, validating our procedures.
|
|
