Umeå University's logo

umu.sePublications
Change search
Link to record
Permanent link

Direct link
Wallin, Erik
Publications (10 of 20) Show all publications
Wallin, E., Wiberg, V. & Servin, M. (2023). Multi-log grasping using reinforcement learning and virtual visual servoing. Robotics, 13(1), Article ID 3.
Open this publication in new window or tab >>Multi-log grasping using reinforcement learning and virtual visual servoing
2023 (English)In: Robotics, E-ISSN 2218-6581, Vol. 13, no 1, article id 3Article in journal (Refereed) Published
Abstract [en]

We explore multi-log grasping using reinforcement learning and virtual visual servoing for automated forwarding in a simulated environment. Automation of forest processes is a major challenge, and many techniques regarding robot control pose different challenges due to the unstructured and harsh outdoor environment. Grasping multiple logs involves various problems of dynamics and path planning, where understanding the interaction between the grapple, logs, terrain, and obstacles requires visual information. To address these challenges, we separate image segmentation from crane control and utilise a virtual camera to provide an image stream from reconstructed 3D data. We use Cartesian control to simplify domain transfer to real-world applications. Because log piles are static, visual servoing using a 3D reconstruction of the pile and its surroundings is equivalent to using real camera data until the point of grasping. This relaxes the limits on computational resources and time for the challenge of image segmentation, and allows for data collection in situations where the log piles are not occluded. The disadvantage is the lack of information during grasping. We demonstrate that this problem is manageable and present an agent that is 95% successful in picking one or several logs from challenging piles of 2–5 logs.

Place, publisher, year, edition, pages
MDPI, 2023
Keywords
autonomous forwarding, visual servoing, virtual camera, reinforcement learning, multi-log grasping, Cartesian control
National Category
Robotics Control Engineering Other Physics Topics Computer Vision and Robotics (Autonomous Systems)
Research subject
Physics; Automatic Control
Identifiers
urn:nbn:se:umu:diva-220402 (URN)10.3390/robotics13010003 (DOI)2-s2.0-85183350095 (Scopus ID)
Projects
Mistra Digital Forest
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, Grant DIA 2017/14 #6
Available from: 2024-02-02 Created: 2024-02-02 Last updated: 2024-02-09Bibliographically approved
Wiberg, V., Wallin, E., Nordjell, T. & Servin, M. (2022). Control of rough terrain vehicles using deep reinforcement learning. IEEE Robotics and Automation Letters, 7(1), 390-397
Open this publication in new window or tab >>Control of rough terrain vehicles using deep reinforcement learning
2022 (English)In: IEEE Robotics and Automation Letters, E-ISSN 2377-3766, Vol. 7, no 1, p. 390-397Article in journal (Refereed) Published
Abstract [en]

We explore the potential to control terrain vehicles using deep reinforcement in scenarios where human operators and traditional control methods are inadequate. This letter presents a controller that perceives, plans, and successfully controls a 16-tonne forestry vehicle with two frame articulation joints, six wheels, and their actively articulated suspensions to traverse rough terrain. The carefully shaped reward signal promotes safe, environmental, and efficient driving, which leads to the emergence of unprecedented driving skills. We test learned skills in a virtual environment, including terrains reconstructed from high-density laser scans of forest sites. The controller displays the ability to handle obstructing obstacles, slopes up to 27 degrees, and a variety of natural terrains, all with limited wheel slip, smooth, and upright traversal with intelligent use of the active suspensions. The results confirm that deep reinforcement learning has the potential to enhance control of vehicles with complex dynamics and high-dimensional observation data compared to human operators or traditional control methods, especially in rough terrain.

Place, publisher, year, edition, pages
IEEE, 2022
Keywords
Artificial Intelligence, Control and Optimization, Computer Science Applications, Computer Vision and Pattern Recognition, Mechanical Engineering, Human-Computer Interaction, Biomedical Engineering, Control and Systems Engineering
National Category
Robotics Computer Vision and Robotics (Autonomous Systems) Other Physics Topics
Research subject
Physics; Computer Science; Automatic Control
Identifiers
urn:nbn:se:umu:diva-189485 (URN)10.1109/lra.2021.3126904 (DOI)000721999500008 ()2-s2.0-85119416792 (Scopus ID)
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, 2017/14 #6eSSENCE - An eScience CollaborationSwedish National Infrastructure for Computing (SNIC), 2021/5-234
Available from: 2021-11-12 Created: 2021-11-12 Last updated: 2024-01-17Bibliographically approved
Wallin, E. & Servin, M. (2022). Data-driven model order reduction for granular media. Computational Particle Mechanics, 9, 15-28
Open this publication in new window or tab >>Data-driven model order reduction for granular media
2022 (English)In: Computational Particle Mechanics, ISSN 2196-4378, Vol. 9, p. 15-28Article in journal (Refereed) Published
Abstract [en]

We investigate the use of reduced-order modelling to run discrete element simulations at higher speeds. Taking a data-drivenapproach, we run many offline simulations in advance and train a model to predict the velocity field from the mass distributionand system control signals. Rapid model inference of particle velocities replaces the intense process of computing contactforces and velocity updates. In coupled DEM and multibody system simulation, the predictor model can be trained to outputthe interfacial reaction forces as well. An adaptive model order reduction technique is investigated, decomposing the mediain domains of solid, liquid, and gaseous state. The model reduction is applied to solid and liquid domains where the particlemotion is strongly correlated with the mean flow, while resolved DEM is used for gaseous domains. Using a ridge regressionpredictor, the performance is tested on simulations of a pile discharge and bulldozing. The measured accuracy is about 90%and 65%, respectively, and the speed-up range between 10 and 60. 

Place, publisher, year, edition, pages
Springer Nature, 2022
National Category
Other Physics Topics Applied Mechanics
Research subject
Physics
Identifiers
urn:nbn:se:umu:diva-169604 (URN)10.1007/s40571-020-00387-6 (DOI)000616428800001 ()2-s2.0-85101460553 (Scopus ID)
Funder
Vinnova, 2019-04832eSSENCE - An eScience Collaboration
Available from: 2020-04-08 Created: 2020-04-08 Last updated: 2023-03-24Bibliographically approved
Wallin, E., Wiberg, V., Vesterlund, F., Holmgren, J., Persson, H. & Servin, M. (2022). Learning multiobjective rough terrain traversability. Journal of terramechanics, 102, 17-26
Open this publication in new window or tab >>Learning multiobjective rough terrain traversability
Show others...
2022 (English)In: Journal of terramechanics, ISSN 0022-4898, E-ISSN 1879-1204, Vol. 102, p. 17-26Article in journal (Refereed) Published
Abstract [en]

We present a method that uses high-resolution topography data of rough terrain, and ground vehicle simulation, to predict traversability. Traversability is expressed as three independent measures: the ability to traverse the terrain at a target speed, energy consumption, and acceleration. The measures are continuous and reflect different objectives for planning that go beyond binary classification. A deep neural network is trained to predict the traversability measures from the local heightmap and target speed. To produce training data, we use an articulated vehicle with wheeled bogie suspensions and procedurally generated terrains. We evaluate the model on laser-scanned forest terrains, previously unseen by the model. The model predicts traversability with an accuracy of 90%. Predictions rely on features from the high-dimensional terrain data that surpass local roughness and slope relative to the heading. Correlations show that the three traversability measures are complementary to each other. With an inference speed 3000 times faster than the ground truth simulation and trivially parallelizable, the model is well suited for traversability analysis and optimal path planning over large areas.

Place, publisher, year, edition, pages
Elsevier, 2022
Keywords
Traversability, Rough terrain vehicle, Multibody simulation, Laser scan, Deep learning
National Category
Robotics
Research subject
Physics; Computerized Image Analysis; data science; Physics
Identifiers
urn:nbn:se:umu:diva-193680 (URN)10.1016/j.jterra.2022.04.002 (DOI)000805039700001 ()2-s2.0-85129503568 (Scopus ID)
Projects
Mistra Digital Forest
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, 2017/14 #6eSSENCE - An eScience CollaborationSwedish National Infrastructure for Computing (SNIC), 2021/5-234
Available from: 2022-04-11 Created: 2022-04-11 Last updated: 2023-05-08Bibliographically approved
Wiberg, V., Wallin, E., Wadbro, E., Rosander, M. & Servin, M. (2022). Simulation-to-reality transfer to control a forwarder with active suspensions through deep reinforcement learning. In: International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA: . Paper presented at International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA, October 4-7, 2022.
Open this publication in new window or tab >>Simulation-to-reality transfer to control a forwarder with active suspensions through deep reinforcement learning
Show others...
2022 (English)In: International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA, 2022Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

Automating the loaded and unloaded driving of a forwarder has the potential to reduce operational costsup to 10% in cut-to-length logging, but remains a challening and unsolved task. The complex interaction between the vehicle and terrain requires the controller to percieve its surroundings and thestate of the vehicle to plan for traversal. Because the state space is high dimensional and the systemdynamics cannot be formulated in closed form or easily approximated, traditional control methods areinadequate. Under these conditions, where learning to act in the environement is easier than learning thesystem dynamics, model free reinforcement learning is a promising option. We use deep reinforcementlearning for control of a 16-tonne forwarder with actively articulated suspensions. To efficiently gathergeneralizable experience, the control policies were safetly trained in simulation while varying severaldomain parameters. Each policy is trained during what correponds to roughly one month of real time.In simulation, the controller shows the ability to traverse rough terrains reconstruced from high-densitylaser scans and handles slopes up to 27◦. To compare the simulated to real performance we transfer thecontrol policies to the physical vehicle. Our results provide insight on how to improve policy transfer toheavy and expensive forest machines.

National Category
Robotics Computer Vision and Robotics (Autonomous Systems) Other Physics Topics
Research subject
Physics
Identifiers
urn:nbn:se:umu:diva-199447 (URN)
Conference
International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA, October 4-7, 2022
Available from: 2022-09-17 Created: 2022-09-17 Last updated: 2022-09-19
Lundbäck, M., Nordfjell, T., Wiberg, V., Wallin, E. & Servin, M. (2022). Traversability analysis using high-resolution laser-scans, simulation, and deep learning. In: : . Paper presented at International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA, October 4-7, 2022.
Open this publication in new window or tab >>Traversability analysis using high-resolution laser-scans, simulation, and deep learning
Show others...
2022 (English)Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

Traversability is of major importance in forestry, where heavy vehicles, weighing up to 40 tons whenfully loaded, traverse rough and sometimes soft terrain. Forest remote sensing is becoming available atresolutions where surface roughness and slope can be determined at length-scales smaller than the forestmachines. Using 3D multibody dynamics simulation of a forest machine driving in virtual terrain replications, the interaction can be captured in great detail. The observed traversability is then automaticallya function of the vehicle geometry, dynamics, and of the local terrain topography relative to heading. Weexpress traversability with three complementary measures: i) the ability to traverse the terrain at a target speed, ii) energy consumption, and iii) machine body acceleration. For high traversability, the lattertwo should be as small as possible while the first measure is at maximum. The simulations are, however,too slow for systematically probing the traversability over large areas. Instead, a deep neural networkis trained to predict the traversability measures from the local heightmap and target speed. The trainingdata comes from simulations of an articulated vehicle with wheeled bogie suspensions driving over procedurally generated terrains while observing the dynamics and local terrain topology. We evaluate themodel on laser-scanned forest terrains, previously unseen by the model. The model predicts traversability with an accuracy of 90% on terrains with 0.25 m resolution and it is 3000 times faster than the groundtruth realtime simulation and trivially parallelizable, making it well suited for traversability analysis andoptimal route planning over large areas. The trained model depends on the vehicle heading, target speed,and detailed features in the topography that a model based only on local slope and roughness cannotcapture. We explore traversability statistics over large areas of laser-scanned terrains and discuss howthe model can be used as a complement or in place of the currently used terrain classification scheme.

National Category
Computer Vision and Robotics (Autonomous Systems) Robotics Other Physics Topics
Research subject
Physics
Identifiers
urn:nbn:se:umu:diva-199448 (URN)
Conference
International Conference of Forest Engineering COFE-FORMEC-IUFRO, Corvallis, Oregon, USA, October 4-7, 2022
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, 2017/14 #6
Available from: 2022-09-17 Created: 2022-09-17 Last updated: 2023-07-03
Servin, M., Vesterlund, F. & Wallin, E. (2021). Digital Twins with Distributed Particle Simulation for Mine-to-Mill Material Tracking. Minerals, 11(5), Article ID 524.
Open this publication in new window or tab >>Digital Twins with Distributed Particle Simulation for Mine-to-Mill Material Tracking
2021 (English)In: Minerals, E-ISSN 2075-163X, Vol. 11, no 5, article id 524Article in journal (Refereed) Published
Abstract [en]

Systems for transport and processing of granular media are challenging to analyse, operate and optimise. In the mining and mineral processing industries, these systems are chains of processes with a complex interplay among the equipment, control and processed material. The material properties have natural variations that are usually only known at certain locations. Therefore, we explored a material-oriented approach to digital twins with a particle representation of the granular media. In digital form, the material is treated as pseudo-particles, each representing a large collection of real particles of various sizes, shapes and mineral properties. Movements and changes in the state of the material are determined by the combined data from control systems, sensors, vehicle telematics and simulation models at locations where no real sensors could see. The particle-based representation enables material tracking along the chain of processes. Each digital particle can act as a carrier of observational data generated by the equipment as it interacts with the real material. This make it possible to better learn the material properties from process observations and to predict the effect on downstream processes. We tested the technique on a mining simulator and demonstrated the analysis that can be performed using data from cross-system material tracking.

Keywords
granular media, mine-to-mill optimisation, material tracking, digital twin, discrete element simulation
National Category
Other Physics Topics Other Computer and Information Science Computational Mathematics Mineral and Mine Engineering Robotics
Research subject
Physics
Identifiers
urn:nbn:se:umu:diva-183071 (URN)10.3390/min11050524 (DOI)000662462200001 ()2-s2.0-85105745382 (Scopus ID)
Funder
Vinnova, 2019-04832eSSENCE - An eScience CollaborationSwedish National Infrastructure for Computing (SNIC), 2018-05973
Available from: 2021-05-15 Created: 2021-05-15 Last updated: 2024-01-17Bibliographically approved
Servin, M., Vesterlund, F. & Wallin, E. (2021). Digital twins with embedded particle simulation. In: Francisco Chinesta, Rémi Abgrall, Olivier Allix, David Néron, Michael Kaliske (Ed.), 14th WCCM & ECCOMAS Congress 2020: Virtual Congress 11 - 15 January, 2021. Paper presented at 14th World Congress in Computational Mechanics (WCCM) ECCOMAS Congress, Virtual, January 11-15, 2021 (pp. 2965-2966). International Centre for Numerical Methods in Engineering (CIMNE)
Open this publication in new window or tab >>Digital twins with embedded particle simulation
2021 (English)In: 14th WCCM & ECCOMAS Congress 2020: Virtual Congress 11 - 15 January, 2021 / [ed] Francisco Chinesta, Rémi Abgrall, Olivier Allix, David Néron, Michael Kaliske, International Centre for Numerical Methods in Engineering (CIMNE) , 2021, p. 2965-2966Conference paper, Oral presentation with published abstract (Other academic)
Place, publisher, year, edition, pages
International Centre for Numerical Methods in Engineering (CIMNE), 2021
Keywords
Digital Twin, Granular Dynamics, Mining, Model Order Reduction
National Category
Other Physics Topics Robotics Other Computer and Information Science
Research subject
Physics
Identifiers
urn:nbn:se:umu:diva-176564 (URN)978-84-121101-7-3 (ISBN)
Conference
14th World Congress in Computational Mechanics (WCCM) ECCOMAS Congress, Virtual, January 11-15, 2021
Available from: 2020-11-06 Created: 2020-11-06 Last updated: 2021-07-27Bibliographically approved
Andersson, J., Bodin, K., Lindmark, D., Servin, M. & Wallin, E. (2021). Reinforcement Learning Control of a Forestry Crane Manipulator. In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021): Proceedings. Paper presented at IROS 2021, IEEE/RSJ International Conference on Intelligent Robots and Systems, Online via Prague, Czech Republic, September 27 - October 1, 2021 (pp. 2121-2126). Prague: IEEE Robotics and Automation Society
Open this publication in new window or tab >>Reinforcement Learning Control of a Forestry Crane Manipulator
Show others...
2021 (English)In: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021): Proceedings, Prague: IEEE Robotics and Automation Society, 2021, p. 2121-2126Conference paper, Published paper (Refereed)
Abstract [en]

Forestry machines are heavy vehicles performing complex manipulation tasks in unstructured production forest environments. Together with the complex dynamics of the on-board hydraulically actuated cranes, the rough forest terrains have posed a particular challenge in forestry automation. In this study, the feasibility of applying reinforcement learning control to forestry crane manipulators is investigated in a simulated environment. Our results show that it is possible to learn successful actuator-space control policies for energy efficient log grasping by invoking a simple curriculum in a deep reinforcement learning setup. Given the pose of the selected logs, our best control policy reaches a grasping success rate of 97%. Including an energy-optimization goal in the reward function, the energy consumption is significantly reduced compared to control policies learned without incentive for energy optimization, while the increase in cycle time is marginal. The energy-optimization effects can be observed in the overall smoother motion and acceleration profiles during crane manipulation. 

Place, publisher, year, edition, pages
Prague: IEEE Robotics and Automation Society, 2021
Keywords
Reinforcement Learning, Robotics and Automation in Agriculture and Forestry, Deep Learning in Grasping and Manipulation
National Category
Control Engineering Robotics Computer Sciences Applied Mechanics
Research subject
Physics; Computer Science
Identifiers
urn:nbn:se:umu:diva-187948 (URN)10.1109/IROS51168.2021.9636219 (DOI)000755125501096 ()2-s2.0-85124340717 (Scopus ID)978-1-6654-1715-0 (ISBN)978-1-6654-1714-3 (ISBN)
Conference
IROS 2021, IEEE/RSJ International Conference on Intelligent Robots and Systems, Online via Prague, Czech Republic, September 27 - October 1, 2021
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, 2017/14 #6
Available from: 2021-09-27 Created: 2021-09-27 Last updated: 2023-08-07Bibliographically approved
Servin, M., Götz, H., Berglund, T. & Wallin, E. (2021). Towards a graph neural network solver for granular dynamics. In: VII International Conference on Particle-based Methods (PARTICLES 2021): Technical program. Paper presented at PARTICLES 2021: VIIth International Conference on Particle-Based Methods, Hamburg, Germany / Hybrid, online, October 4-5, 2021. Hamburg, Germany
Open this publication in new window or tab >>Towards a graph neural network solver for granular dynamics
2021 (English)In: VII International Conference on Particle-based Methods (PARTICLES 2021): Technical program, Hamburg, Germany, 2021Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

The discrete element method (DEM) is a versatile but computationally intensive method for granular dynamics simulation. We investigate the possibility of accelerating DEM simulations using graph neural networks (GNN), which automatically support variable connectivity between particles. This approach was recently found promising for particle-based simulation of complex fluids [1]. We start from a time-implicit, or nonsmooth, DEM [2], where the computational bottleneck is the process of solving a mixed linear complementarity problem (MLCP) to obtain the contact forces and particle velocity update. This solve step is substituted by a GNN, trained to predict the MLCP solution. Following [1], we employ an encoder-process-decoder structure for the GNN. The particle and connectivity data is encoded in an input graph with particle mass, external force, and previous velocity as node attributes, and contact overlap, normal, and tangent vectors as edge attributes. The sought solution is represented in the output graph with the updated particle velocities as node attributes and the contact forces as edge attributes. In the intermediate processing step, the input graph is converted to a latent graph, which is then advanced with a fixed number of message passing steps involving a multilayer perceptron neural network for updating the edge and node values. The output graph, with the approximate solution to the MLCP, is finally computed by decoding the last processed latent graph.

Both a supervised and unsupervised method are tested for training the network on granular simulation of particles in a rotating or static drum. AGX Dynamics [3] is used for running the simulations, and Pytorch [4] in combination with the Deep Graph Library [5] for the learning. The supervised model learns from ground truth MLCP solutions, computed using a projected Gauss-Seidel (PGS) solver, sampled from 1200 simulations involving 50-150 particles. The unsupervised model learns to minimize a loss function derived from the MLCP residual function using particle configurations extracted from the same simulations but ignoring the approximate solution from the PGS solver. The simulation samples are split into training data (80%), validation data (10%), and test data (10%). Network hyperparameter optimization is performed. The supervised GNN solver reaches an error level of 1% for the contact forces and 0.01% on the particle velocities for a static drum. For a rotating drum, the respective errors are 10% and 1%. The unsupervised GNN solver reaches 1% velocity errors, 5% normal forces errors, but it has significant problems with predicting the friction forces. The latter is presumably because of the discontinuous loss function that follows from the Coulomb friction law and therefore we explore regularization of it. Finally, we discuss the potential scalability and performance for large particle systems.

Place, publisher, year, edition, pages
Hamburg, Germany: , 2021
National Category
Physical Sciences Computer Sciences Fluid Mechanics and Acoustics
Research subject
Physics; Computer Science
Identifiers
urn:nbn:se:umu:diva-187950 (URN)
Conference
PARTICLES 2021: VIIth International Conference on Particle-Based Methods, Hamburg, Germany / Hybrid, online, October 4-5, 2021
Funder
Mistra - The Swedish Foundation for Strategic Environmental Research, 2017/14 #6
Available from: 2021-09-27 Created: 2021-09-27 Last updated: 2021-11-29Bibliographically approved
Organisations

Search in DiVA

Show all publications