Latent Action Priors From a Single Gait Cycle Demonstration for Online Imitation Learning

Deep Reinforcement Learning (DRL) in simulation often results in brittle and unrealistic learning outcomes. To push the agent towards more desirable solutions, prior information can be injected in the learning process through, for instance, reward shaping, expert data, or motion primitives. We propose an additional inductive bias for robot learning: latent actions learned from expert demonstration as priors in the action space. We show that these action priors can be learned from only a single open-loop gait cycle using a simple autoencoder. Using these latent action priors combined with established style rewards for imitation in DRL achieves above expert demonstration level of performance and leads to more desirable gaits. Further, action priors substantially improve the performance on transfer tasks, even leading to gait transitions for higher target speeds. Videos and code are available at https://sites.google.com/view/latent-action-priors.

Experience Transfer for Robust Direct Data-Driven Control

Learning-based control uses data to design efficient controllers for specific systems. When multiple systems are involved, experience transfer usually focuses on data availability and controller performance yet neglects robustness to variations between systems. In contrast, this letter explores experience transfer from a robustness perspective. We leverage the transfer to design controllers that are robust not only to the uncertainty regarding an individual agent's model but also to the choice of agent in a fleet. Experience transfer enables the design of safe and robust controllers that work out of the box for all systems in a heterogeneous fleet. Our approach combines scenario optimization and recent formulations for direct data-driven control without the need to estimate a model of the system or determine uncertainty bounds for its parameters. We demonstrate the benefits of our data-driven robustification method through a numerical case study and obtain learned controllers that generalize well from a small number of open-loop trajectories in a quadcopter simulation.

Multi-Arm Bin-Picking in Real-Time: A Combined Task and Motion Planning Approach

Automated bin-picking is a prerequisite for fully automated manufacturing and warehouses. To successfully pick an item from an unstructured bin the robot needs to first detect possible grasps for the objects, decide on the object to remove and consequently plan and execute a feasible trajectory to retrieve the chosen object. Over the last years significant progress has been made towards solving these problems. However, when multiple robot arms are cooperating the decision and planning problems become exponentially harder. We propose an integrated multi-arm bin-picking pipeline (IMAPIP), and demonstrate that it is able to reliably pick objects from a bin in real-time using multiple robot arms. IMAPIP solves the multi-arm bin-picking task first at high-level using a geometry-aware policy integrated in a combined task and motion planning framework. We then plan motions consistent with this policy using the BIT* algorithm on the motion planning level. We show that this integrated solution enables robot arm cooperation. In our experiments, we show the proposed geometry-aware policy outperforms a baseline by increasing bin-picking time by 28\% using two robot arms. The policy is robust to changes in the position of the bin and number of objects. We also show that IMAPIP to successfully scale up to four robot arms working in close proximity.

Event-Triggered Time-Varying Bayesian Optimization

We consider the problem of sequentially optimizing a time-varying objective function using time-varying Bayesian optimization (TVBO). Here, the key challenge is to cope with old data. Current approaches to TVBO require prior knowledge of a constant rate of change. However, the rate of change is usually neither known nor constant. We propose an event-triggered algorithm, ET-GP-UCB, that detects changes in the objective function online. The event-trigger is based on probabilistic uniform error bounds used in Gaussian process regression. The trigger automatically detects when significant change in the objective functions occurs. The algorithm then adapts to the temporal change by resetting the accumulated dataset. We provide regret bounds for ET-GP-UCB and show in numerical experiments that it is competitive with state-of-the-art algorithms even though it requires no knowledge about the temporal changes. Further, ET-GP-UCB outperforms these competitive baselines if the rate of change is misspecified and we demonstrate that it is readily applicable to various settings without tuning hyperparameters.

Improving the Performance of Robust Control through Event-Triggered Learning

Robust controllers ensure stability in feedback loops designed under uncertainty but at the cost of performance. Model uncertainty in time-invariant systems can be reduced by recently proposed learning-based methods, thus improving the performance of robust controllers using data. However, in practice, many systems also exhibit uncertainty in the form of changes over time, e.g., due to weight shifts or wear and tear, leading to decreased performance or instability of the learning-based controller. We propose an event-triggered learning algorithm that decides when to learn in the face of uncertainty in the LQR problem with rare or slow changes. Our key idea is to switch between robust and learned controllers. For learning, we first approximate the optimal length of the learning phase via Monte-Carlo estimations using a probabilistic model. We then design a statistical test for uncertain systems based on the moment-generating function of the LQR cost. The test detects changes in the system under control and triggers re-learning when control performance deteriorates due to system changes. We demonstrate improved performance over a robust controller baseline in a numerical example.

On Controller Tuning with Time-Varying Bayesian Optimization

Changing conditions or environments can cause system dynamics to vary over time. To ensure optimal control performance, controllers should adapt to these changes. When the underlying cause and time of change is unknown, we need to rely on online data for this adaptation. In this paper, we will use time-varying Bayesian optimization (TVBO) to tune controllers online in changing environments using appropriate prior knowledge on the control objective and its changes. Two properties are characteristic of many online controller tuning problems: First, they exhibit incremental and lasting changes in the objective due to changes to the system dynamics, e.g., through wear and tear. Second, the optimization problem is convex in the tuning parameters. Current TVBO methods do not explicitly account for these properties, resulting in poor tuning performance and many unstable controllers through over-exploration of the parameter space. We propose a novel TVBO forgetting strategy using Uncertainty-Injection (UI), which incorporates the assumption of incremental and lasting changes. The control objective is modeled as a spatio-temporal Gaussian process (GP) with UI through a Wiener process in the temporal domain. Further, we explicitly model the convexity assumptions in the spatial dimension through GP models with linear inequality constraints. In numerical experiments, we show that our model outperforms the state-of-the-art method in TVBO, exhibiting reduced regret and fewer unstable parameter configurations.

Local policy search with Bayesian optimization

Reinforcement learning (RL) aims to find an optimal policy by interaction with an environment. Consequently, learning complex behavior requires a vast number of samples, which can be prohibitive in practice. Nevertheless, instead of systematically reasoning and actively choosing informative samples, policy gradients for local search are often obtained from random perturbations. These random samples yield high variance estimates and hence are sub-optimal in terms of sample complexity. Actively selecting informative samples is at the core of Bayesian optimization, which constructs a probabilistic surrogate of the objective from past samples to reason about informative subsequent ones. In this paper, we propose to join both worlds. We develop an algorithm utilizing a probabilistic model of the objective function and its gradient. Based on the model, the algorithm decides where to query a noisy zeroth-order oracle to improve the gradient estimates. The resulting algorithm is a novel type of policy search method, which we compare to existing black-box algorithms. The comparison reveals improved sample complexity and reduced variance in extensive empirical evaluations on synthetic objectives. Further, we highlight the benefits of active sampling on popular RL benchmarks.

Probabilistic robust linear quadratic regulators with Gaussian processes

Probabilistic models such as Gaussian processes (GPs) are powerful tools to learn unknown dynamical systems from data for subsequent use in control design. While learning-based control has the potential to yield superior performance in demanding applications, robustness to uncertainty remains an important challenge. Since Bayesian methods quantify uncertainty of the learning results, it is natural to incorporate these uncertainties into a robust design. In contrast to most state-of-the-art approaches that consider worst-case estimates, we leverage the learning method's posterior distribution in the controller synthesis. The result is a more informed and, thus, more efficient trade-off between performance and robustness. We present a novel controller synthesis for linearized GP dynamics that yields robust controllers with respect to a probabilistic stability margin. The formulation is based on a recently proposed algorithm for linear quadratic control synthesis, which we extend by giving probabilistic robustness guarantees in the form of credibility bounds for the system's stability.Comparisons to existing methods based on worst-case and certainty-equivalence designs reveal superior performance and robustness properties of the proposed method.

Excursion Search for Constrained Bayesian Optimization under a Limited Budget of Failures

When learning to ride a bike, a child falls down a number of times before achieving the first success. As falling down usually has only mild consequences, it can be seen as a tolerable failure in exchange for a faster learning process, as it provides rich information about an undesired behavior. In the context of Bayesian optimization under unknown constraints (BOC), typical strategies for safe learning explore conservatively and avoid failures by all means. On the other side of the spectrum, non conservative BOC algorithms that allow failing may fail an unbounded number of times before reaching the optimum. In this work, we propose a novel decision maker grounded in control theory that controls the amount of risk we allow in the search as a function of a given budget of failures. Empirical validation shows that our algorithm uses the failures budget more efficiently in a variety of optimization experiments, and generally achieves lower regret, than state-of-the-art methods. In addition, we propose an original algorithm for unconstrained Bayesian optimization inspired by the notion of excursion sets in stochastic processes, upon which the failures-aware algorithm is built.

A Learnable Safety Measure

Failures are challenging for learning to control physical systems since they risk damage, time-consuming resets, and often provide little gradient information. Adding safety constraints to exploration typically requires a lot of prior knowledge and domain expertise. We present a safety measure which implicitly captures how the system dynamics relate to a set of failure states. Not only can this measure be used as a safety function, but also to directly compute the set of safe state-action pairs. Further, we show a model-free approach to learn this measure by active sampling using Gaussian processes. While safety can only be guaranteed after learning the safety measure, we show that failures can already be greatly reduced by using the estimated measure during learning.