Multi-object grasping is a challenging task. It is important for energy and cost-efficient operation of industrial crane manipulators, such as those used to collect tree logs off the forest floor and onto forest machines. In this work, we used synthetic data from physics simulations to explore how data-driven modeling can be used to infer multi-object grasp poses from images. We showed that convolutional neural networks can be trained specifically for synthesizing multi-object grasps. Using RGB-Depth images and instance segmentation masks as input, a U-Net model outputs grasp maps with corresponding grapple orientation and opening width. Given an observation of a pile of logs, the model can be used to synthesize and rate the possible grasp poses and select the most suitable one, with the possibility to respect changing operational constraints such as lift capacity and reach. When tested on previously unseen data, the proposed model found successful grasp poses with an accuracy of 95%.
We explore multi-log grasping using reinforcement learning and virtual visual servoing for automated forwarding. 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 problems of dynamics and path planning, where the interaction between the grapple, logs, terrain, and obstacles requires visual information. To address these challenges, we separate image segmentation from crane control and utilize a virtual camera to provide an image stream from 3D reconstructed data. We use Cartesian control to simplify domain transfer. 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 limit 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.
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 at nearly 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 look-ahead planning. However, the strong transfer capabilities entail that further development concerning perception and performance can be largely confined to simulation.
Open-set semi-supervised learning (OSSL) is a realistic setting of semi-supervised learning where the unlabeled training set contains classes that are not present in the labeled set. Many existing OSSL methods assume that these out-of-distribution data are harmful and put effort into excluding data from unknown classes from the training objective. In contrast, we propose an OSSL framework that facilitates learning from all unlabeled data through self-supervision. Additionally, we utilize an energy-based score to accurately recognize data belonging to the known classes, making our method well-suited for handling uncurated data in deployment. We show through extensive experimental evaluations on several datasets that our method shows overall unmatched robustness and performance in terms of closed-set accuracy and open-set recognition compared with state-of-the-art for OSSL. Our code will be released upon publication.
Following the success of supervised learning, semi-supervised learning (SSL) is now becoming increasingly popular. SSL is a family of methods, which in addition to a labeled training set, also use a sizable collection of unlabeled data for fitting a model. Most of the recent successful SSL methods are based on pseudo-labeling approaches: letting confident model predictions act as training labels. While these methods have shown impressive results on many benchmark datasets, a drawback of this approach is that not all unlabeled data are used during training. We propose a new SSL algorithm, DoubleMatch, which combines the pseudo-labeling technique with a self-supervised loss, enabling the model to utilize all unlabeled data in the training process. We show that this method achieves state-of-the-art accuracies on multiple benchmark datasets while also reducing training times compared to existing SSL methods. Code is available at https://github.com/walline/doublematch.
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.
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.
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.