With the ever increasing number of active satellites in space, the rising demand for larger formations of small satellites and the commercialization of the space industry (so-called New Space), the realization of manufacturing processes in orbit comes closer to reality. Reducing launch costs and risks, allowing for faster on-demand deployment of individually configured satellites as well as the prospect for possible on-orbit servicing for satellites makes the idea of realizing an in-orbit factory promising. In this paper, we present a novel approach to an in-orbit factory of small satellites covering a digital process twin, AI-based fault detection, and teleoperated robot-control, which are being researched as part of the "AI-enabled Cyber-Physical In-Orbit Factory" project. In addition to the integration of modern automation and Industry 4.0 production approaches, the question of how artificial intelligence (AI) and learning approaches can be used to make the production process more robust, fault-tolerant and autonomous is addressed. This lays the foundation for a later realisation of satellite production in space in the form of an in-orbit factory. Central aspect is the development of a robotic AIT (Assembly, Integration and Testing) system where a small satellite could be assembled by a manipulator robot from modular subsystems. Approaches developed to improving this production process with AI include employing neural networks for optical and electrical fault detection of components. Force sensitive measuring and motion training helps to deal with uncertainties and tolerances during assembly. An AI-guided teleoperated control of the robot arm allows for human intervention while a Digital Process Twin represents process data and provides supervision during the whole production process. Approaches and results towards automated satellite production are presented in detail.
Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train generalist X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. More details can be found on the project website $\href{https://robotics-transformer-x.github.io}{\text{robotics-transformer-x.github.io}}$.
In search of the simplest baseline capable of competing with Deep Reinforcement Learning on locomotion tasks, we propose a biologically inspired model-free open-loop strategy. Drawing upon prior knowledge and harnessing the elegance of simple oscillators to generate periodic joint motions, it achieves respectable performance in five different locomotion environments, with a number of tunable parameters that is a tiny fraction of the thousands typically required by RL algorithms. Unlike RL methods, which are prone to performance degradation when exposed to sensor noise or failure, our open-loop oscillators exhibit remarkable robustness due to their lack of reliance on sensors. Furthermore, we showcase a successful transfer from simulation to reality using an elastic quadruped, all without the need for randomization or reward engineering.
Spring-based actuators in legged locomotion provide energy-efficiency and improved performance, but increase the difficulty of controller design. Whereas previous works have focused on extensive modeling and simulation to find optimal controllers for such systems, we propose to learn model-free controllers directly on the real robot. In our approach, gaits are first synthesized by central pattern generators (CPGs), whose parameters are optimized to quickly obtain an open-loop controller that achieves efficient locomotion. Then, to make that controller more robust and further improve the performance, we use reinforcement learning to close the loop, to learn corrective actions on top of the CPGs. We evaluate the proposed approach in DLR's elastic quadruped bert. Our results in learning trotting and pronking gaits show that exploitation of the spring actuator dynamics emerges naturally from optimizing for dynamic motions, yielding high-performing locomotion despite being model-free. The whole process takes no more than 1.5 hours on the real robot and results in natural-looking gaits.
Reinforcement learning (RL) enables robots to learn skills from interactions with the real world. In practice, the unstructured step-based exploration used in Deep RL -- often very successful in simulation -- leads to jerky motion patterns on real robots. Consequences of the resulting shaky behavior are poor exploration, or even damage to the robot. We address these issues by adapting state-dependent exploration (SDE) to current Deep RL algorithms. To enable this adaptation, we propose three extensions to the original SDE, which leads to a new exploration method generalized state-dependent exploration (gSDE). We evaluate gSDE both in simulation, on PyBullet continuous control tasks, and directly on a tendon-driven elastic robot. gSDE yields competitive results in simulation but outperforms the unstructured exploration on the real robot. The code is available at https://github.com/DLR-RM/stable-baselines3/tree/sde.
Physical modeling of robotic system behavior is the foundation for controlling many robotic mechanisms to a satisfactory degree. Mechanisms are also typically designed in a way that good model accuracy can be achieved with relatively simple models and model identification strategies. If the modeling accuracy using physically based models is not enough or too complex, model-free methods based on machine learning techniques can help. Of particular interest to us was therefore the question to what degree semi-parametric modeling techniques, meaning combinations of physical models with machine learning, increase the modeling accuracy of inverse dynamics models which are typically used in robot control. To this end, we evaluated semi-parametric Gaussian process regression and a novel model-based neural network architecture, and compared their modeling accuracy to a series of naive semi-parametric, parametric-only and non-parametric-only regression methods. The comparison has been carried out on three test scenarios, one involving a real test-bed and two involving simulated scenarios, with the most complex scenario targeting the modeling a simulated robot's inverse dynamics model. We found that in all but one case, semi-parametric Gaussian process regression yields the most accurate models, also with little tuning required for the training procedure.
Deep Reinforcement Learning has shown great success in a variety of control tasks. However, it is unclear how close we are to the vision of putting Deep RL into practice to solve real world problems. In particular, common practice in the field is to train policies on largely deterministic simulators and to evaluate algorithms through training performance alone, without a train/test distinction to ensure models generalise and are not overfitted. Moreover, it is not standard practice to check for generalisation under domain shift, although robustness to such system change between training and testing would be necessary for real-world Deep RL control, for example, in robotics. In this paper we study these issues by first characterising the sources of uncertainty that provide generalisation challenges in Deep RL. We then provide a new benchmark and thorough empirical evaluation of generalisation challenges for state of the art Deep RL methods. In particular, we show that, if generalisation is the goal, then common practice of evaluating algorithms based on their training performance leads to the wrong conclusions about algorithm choice. Finally, we evaluate several techniques for improving generalisation and draw conclusions about the most robust techniques to date.
Continuous action policy search, the search for efficient policies in continuous control tasks, is currently the focus of intensive research driven both by the recent success of deep reinforcement learning algorithms and by the emergence of competitors based on evolutionary algorithms. In this paper, we present a broad survey of policy search methods, incorporating into a common big picture these very different approaches as well as alternatives such as Bayesian Optimization and directed exploration methods. The main message of this overview is in the relationship between the families of methods, but we also outline some factors underlying sample efficiency properties of the various approaches. Besides, to keep this survey as short and didactic as possible, we do not go into the details of mathematical derivations of the elementary algorithms.