Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
7
result(s) for
"Wiberg, Viktor"
Sort by:
Multi-Log Grasping Using Reinforcement Learning and Virtual Visual Servoing
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.
Journal Article
Multi-log grasping using reinforcement learning and virtual visual servoing
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. Since 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 collecting data 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.
Paper
Sim-to-real transfer of active suspension control using deep reinforcement learning
We explore sim-to-real transfer of deep reinforcement learning controllers for a heavy vehicle with active suspensions designed for traversing rough terrain. While related research primarily focuses on lightweight robots with electric motors and fast actuation, this study uses a forestry vehicle with a complex hydraulic driveline and slow actuation. We simulate the vehicle using multibody dynamics and apply system identification to find an appropriate set of simulation parameters. We then train policies in simulation using various techniques to mitigate the sim-to-real gap, including domain randomization, action delays, and a reward penalty to encourage smooth control. In reality, the policies trained with action delays and a penalty for erratic actions perform nearly at the same level as in simulation. In experiments on level ground, the motion trajectories closely overlap when turning to either side, as well as in a route tracking scenario. When faced with a ramp that requires active use of the suspensions, the simulated and real motions are in close alignment. This shows that the actuator model together with system identification yields a sufficiently accurate model of the actuators. We observe that policies trained without the additional action penalty exhibit fast switching or bang-bang control. These present smooth motions and high performance in simulation but transfer poorly to reality. We find that policies make marginal use of the local height map for perception, showing no indications of predictive planning. However, the strong transfer capabilities entail that further development concerning perception and performance can be largely confined to simulation.
Paper
Control of rough terrain vehicles using deep reinforcement learning
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\\(^\\circ\\), 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.
Paper
Learning multiobjective rough terrain traversability
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.
Paper
Computation of value -at -risk: The fast convolution method, dimension reduction and perturbation theory
Value-at-risk is a measure of market risk for a portfolio. Market risk is the chance that the portfolio declines in value due to changes in market variables. This thesis is about the computation of value-at-risk for portfolios with derivatives and for models for returns that have a distribution with fat tails. We introduce a new Fourier algorithm, the fast convolution method, for computing value-at-risk. The fast convolution method is different from other Fourier methods in that it does not require that the characteristic function of the portfolio returns be known explicitly. Our new method can therefore be used with more general return models. In the thesis we present experiments with three return models: the normal model, the asymmetric T model and a model using the non-parametric Parzen density estimator. We also discuss how the fast convolution method can be extended to compute the value-at-risk gradient, present a proof of convergence and illustrate the performance of the method with examples. We develop and compare two methods for dimension reduction in the computation of value-at-risk. The goal of dimension reduction is to reduce computation time by finding a small model that captures the main dynamics of the original model. We compare the two methods for an example problem and conclude that the method based on mean square error is superior. Finally, we present an optimization example that illustrates that dimension reduction may reduce the time to compute value-at-risk while maintaining good accuracy. We develop a perturbation theory for value-at-risk with respect to changes in the return model. By considering variational properties, we derive a first-order error bound and find the condition number of value-at-risk. We argue that the sensitivity observed in empirical studies is an inherent limitation of value-at-risk.
Dissertation