Learning from demonstrations in the wild (e.g. YouTube videos) is a tantalizing goal in imitation learning. However, for this goal to be achieved, imitation learning algorithms must deal with the fact that the demonstrators and learners may have bodies that differ from one another. This condition -- "embodiment mismatch" -- is ignored by many recent imitation learning algorithms. Our proposed imitation learning technique, SILEM (\textbf{S}keletal feature compensation for \textbf{I}mitation \textbf{L}earning with \textbf{E}mbodiment \textbf{M}ismatch), addresses a particular type of embodiment mismatch by introducing a learned affine transform to compensate for differences in the skeletal features obtained from the learner and expert. We create toy domains based on PyBullet's HalfCheetah and Ant to assess SILEM's benefits for this type of embodiment mismatch. We also provide qualitative and quantitative results on more realistic problems -- teaching simulated humanoid agents, including Atlas from Boston Dynamics, to walk by observing human demonstrations.
Chore division is a class of fair division problems in which some undesirable "resource" must be shared among a set of participants, with each participant wanting to get as little as possible. Typically the set of participants is fixed and known at the outset. This paper introduces a novel variant, called sequential online chore division (SOCD), in which participants arrive and depart online, while the chore is being performed: both the total number of participants and their arrival/departure times are initially unknown. In SOCD, exactly one agent must be performing the chore at any give time (e.g. keeping lookout), and switching the performer incurs a cost. In this paper, we propose and analyze three mechanisms for SOCD: one centralized mechanism using side payments, and two distributed ones that seek to balance the participants' loads. Analysis and results are presented in a domain motivated by autonomous vehicle convoy formation, where the chore is leading the convoy so that all followers can enjoy reduced wind resistance.
In imitation learning from observation IfO, a learning agent seeks to imitate a demonstrating agent using only observations of the demonstrated behavior without access to the control signals generated by the demonstrator. Recent methods based on adversarial imitation learning have led to state-of-the-art performance on IfO problems, but they typically suffer from high sample complexity due to a reliance on data-inefficient, model-free reinforcement learning algorithms. This issue makes them impractical to deploy in real-world settings, where gathering samples can incur high costs in terms of time, energy, and risk. In this work, we hypothesize that we can incorporate ideas from model-based reinforcement learning with adversarial methods for IfO in order to increase the data efficiency of these methods without sacrificing performance. Specifically, we consider time-varying linear Gaussian policies, and propose a method that integrates the linear-quadratic regulator with path integral policy improvement into an existing adversarial IfO framework. The result is a more data-efficient IfO algorithm with better performance, which we show empirically in four simulation domains: using far fewer interactions with the environment, the proposed method exhibits similar or better performance than the existing technique.
Creating robots that can perform general-purpose service tasks in a human-populated environment has been a longstanding grand challenge for AI and Robotics research. One particularly valuable skill that is relevant to a wide variety of tasks is the ability to locate and retrieve objects upon request. This paper models this skill as a Scavenger Hunt (SH) game, which we formulate as a variation of the NP-hard stochastic traveling purchaser problem. In this problem, the goal is to find a set of objects as quickly as possible, given probability distributions of where they may be found. We investigate the performance of several solution algorithms for the SH problem, both in simulation and on a real mobile robot. We use Reinforcement Learning (RL) to train an agent to plan a minimal cost path, and show that the RL agent can outperform a range of heuristic algorithms, achieving near optimal performance. In order to stimulate research on this problem, we introduce a publicly available software stack and associated website that enable users to upload scavenger hunts which robots can download, perform, and learn from to continually improve their performance on future hunts.
A desirable goal for autonomous agents is to be able to coordinate on the fly with previously unknown teammates. Known as "ad hoc teamwork", enabling such a capability has been receiving increasing attention in the research community. One of the central challenges in ad hoc teamwork is quickly recognizing the current plans of other agents and planning accordingly. In this paper, we focus on the scenario in which teammates can communicate with one another, but only at a cost. Thus, they must carefully balance plan recognition based on observations vs. that based on communication. This paper proposes a new metric for evaluating how similar are two policies that a teammate may be following - the Expected Divergence Point (EDP). We then present a novel planning algorithm for ad hoc teamwork, determining which query to ask and planning accordingly. We demonstrate the effectiveness of this algorithm in a range of increasingly general communication in ad hoc teamwork problems.
Traffic congestion is a major challenge in modern urban settings. The industry-wide development of autonomous and automated vehicles (AVs) motivates the question of how can AVs contribute to congestion reduction. Past research has shown that in small scale mixed traffic scenarios with both AVs and human-driven vehicles, a small fraction of AVs executing a controlled multiagent driving policy can mitigate congestion. In this paper, we scale up existing approaches and develop new multiagent driving policies for AVs in scenarios with greater complexity. We start by showing that a congestion metric used by past research is manipulable in open road network scenarios where vehicles dynamically join and leave the road. We then propose using a different metric that is robust to manipulation and reflects open network traffic efficiency. Next, we propose a modular transfer reinforcement learning approach, and use it to scale up a multiagent driving policy to outperform human-like traffic and existing approaches in a simulated realistic scenario, which is an order of magnitude larger than past scenarios (hundreds instead of tens of vehicles). Additionally, our modular transfer learning approach saves up to 80% of the training time in our experiments, by focusing its data collection on key locations in the network. Finally, we show for the first time a distributed multiagent policy that improves congestion over human-driven traffic. The distributed approach is more realistic and practical, as it relies solely on existing sensing and actuation capabilities, and does not require adding new communication infrastructure.
This paper presents a learning-based approach to consider the effect of unobservable world states in kinodynamic motion planning in order to enable accurate high-speed off-road navigation on unstructured terrain. Existing kinodynamic motion planners either operate in structured and homogeneous environments and thus do not need to explicitly account for terrain-vehicle interaction, or assume a set of discrete terrain classes. However, when operating on unstructured terrain, especially at high speeds, even small variations in the environment will be magnified and cause inaccurate plan execution. In this paper, to capture the complex kinodynamic model and mathematically unknown world state, we learn a kinodynamic planner in a data-driven manner with onboard inertial observations. Our approach is tested on a physical robot in different indoor and outdoor environments, enables fast and accurate off-road navigation, and outperforms environment-independent alternatives, demonstrating 52.4% to 86.9% improvement in terms of plan execution success rate while traveling at high speeds.
We propose firefly neural architecture descent, a general framework for progressively and dynamically growing neural networks to jointly optimize the networks' parameters and architectures. Our method works in a steepest descent fashion, which iteratively finds the best network within a functional neighborhood of the original network that includes a diverse set of candidate network structures. By using Taylor approximation, the optimal network structure in the neighborhood can be found with a greedy selection procedure. We show that firefly descent can flexibly grow networks both wider and deeper, and can be applied to learn accurate but resource-efficient neural architectures that avoid catastrophic forgetting in continual learning. Empirically, firefly descent achieves promising results on both neural architecture search and continual learning. In particular, on a challenging continual image classification task, it learns networks that are smaller in size but have higher average accuracy than those learned by the state-of-the-art methods.