Robotics (cs.RO)

Abstract: Rigid robots can be precise in repetitive tasks but struggle in unstructured environments. Nature's versatility in such environments inspires researchers to develop biomimetic robots that incorporate compliant and contracting artificial muscles. Among the recently proposed artificial muscle technologies, electrohydraulic actuators are promising since they offer comparable performance to mammalian muscles in terms of speed and power density. However, they require high driving voltages and have safety concerns due to exposed electrodes. These high voltages lead to either bulky or inefficient driving electronics that make untethered, high-degree-of-freedom bio-inspired robots difficult to realize. Here, we present low voltage electrohydraulic actuators (LEAs) that match mammalian skeletal muscles in average power density (50.5 W/kg) and peak strain rate (971 percent/s) at a driving voltage of just 1100 V. This driving voltage is approx. 5 - 7 times lower compared to other electrohydraulic actuators using paraelectric dielectrics. Furthermore, LEAs are safe to touch, waterproof, and self-clearing, which makes them easy to implement in wearables and robotics. We characterize, model, and physically validate key performance metrics of the actuator and compare its performance to state-of-the-art electrohydraulic designs. Finally, we demonstrate the utility of our actuators on two muscle-based electrohydraulic robots: an untethered soft robotic swimmer and a robotic gripper. We foresee that LEAs can become a key building block for future highly-biomimetic untethered robots and wearables with many independent artificial muscles such as biomimetic hands, faces, or exoskeletons.

Abstract: Predicting the future motion of observed vehicles is a crucial enabler for safe autonomous driving. The field of motion prediction has seen large progress recently with State-of-the-Art (SotA) models achieving impressive results on large-scale public benchmarks. However, recent work revealed that learning-based methods are prone to predict off-road trajectories in challenging scenarios. These can be created by perturbing existing scenarios with additional turns in front of the target vehicle while the motion history is left unchanged. We argue that this indicates that SotA models do not consider the map information sufficiently and demonstrate how this can be solved, by representing model inputs and outputs in a Frenet frame defined by lane centreline sequences. To this end, we present a general wrapper that leverages a Frenet representation of the scene and that can be applied to SotA models without changing their architecture. We demonstrate the effectiveness of this approach in a comprehensive benchmark using two SotA motion prediction models. Our experiments show that this reduces the off-road rate on challenging scenarios by more than 90\%, without sacrificing average performance.

Abstract: Motion planning (MP) is one of the core robotics problems requiring fast methods for finding a collision-free robot motion path connecting the given start and goal states. Neural motion planners (NMPs) demonstrate fast computational speed in finding path solutions but require a huge amount of expert trajectories for learning, thus adding a significant training computational load. In contrast, recent advancements have also led to a physics-informed NMP approach that directly solves the Eikonal equation for motion planning and does not require expert demonstrations for learning. However, experiments show that the physics-informed NMP approach performs poorly in complex environments and lacks scalability in multiple scenarios and high-dimensional real robot settings. To overcome these limitations, this paper presents a novel and tractable Eikonal equation formulation and introduces a new progressive learning strategy to train neural networks without expert data in complex, cluttered, multiple high-dimensional robot motion planning scenarios. The results demonstrate that our method outperforms state-of-the-art traditional MP, data-driven NMP, and physics-informed NMP methods by a significant margin in terms of computational planning speed, path quality, and success rates. We also show that our approach scales to multiple complex, cluttered scenarios and the real robot set up in a narrow passage environment. The proposed method's videos and code implementations are available at https://github.com/ruiqini/P-NTFields.

Abstract: One of the motivations for exploring flapping-wing aerial robotic systems is to seek energy reduction, by maintaining manoeuvrability, compared to conventional unmanned aerial systems. A Flapping Wing Flying Robot (FWFR) can glide in favourable wind conditions, decreasing energy consumption significantly. In addition, it is also necessary to investigate the power consumption of the components in the flapping-wing robot. In this work, two sets of the FWFR components are analyzed in terms of power consumption: a) motor/electronics components and b) a vision system for monitoring the environment during the flight. A measurement device is used to record the power utilization of the motors in the launching and ascending phases of the flight and also in cruising flight around the desired height. Additionally, an analysis of event cameras and stereo vision systems in terms of energy consumption has been performed. The results provide a first step towards decreasing battery usage and, consequently, providing additional flight time.

Abstract: Motion planning is integral to robotics applications such as autonomous driving, surgical robots, and industrial manipulators. Existing planning methods lack scalability to higher-dimensional spaces, while recent learning based planners have shown promise in accelerating sampling-based motion planners (SMP) but lack generalizability to out-of-distribution environments. To address this, we present a novel approach, Vector Quantized-Motion Planning Transformers (VQ-MPT) that overcomes the key generalization and scaling drawbacks of previous learning-based methods. VQ-MPT consists of two stages. Stage 1 is a Vector Quantized-Variational AutoEncoder model that learns to represent the planning space using a finite number of sampling distributions, and stage 2 is an Auto-Regressive model that constructs a sampling region for SMPs by selecting from the learned sampling distribution sets. By splitting large planning spaces into discrete sets and selectively choosing the sampling regions, our planner pairs well with out-of-the-box SMPs, generating near-optimal paths faster than without VQ-MPT's aid. It is generalizable in that it can be applied to systems of varying complexities, from 2D planar to 14D bi-manual robots with diverse environment representations, including costmaps and point clouds. Trained VQ-MPT models generalize to environments unseen during training and achieve higher success rates than previous methods.

Abstract: Existing object pose estimation methods commonly require a one-to-one point matching step that forces them to be separated into two consecutive stages: visual correspondence detection (e.g., by matching feature descriptors as part of a perception front-end) followed by geometric alignment (e.g., by optimizing a robust estimation objective for pointcloud registration or perspective-n-point). Instead, we propose a matching-free probabilistic formulation with two main benefits: i) it enables unified and concurrent optimization of both visual correspondence and geometric alignment, and ii) it can represent different plausible modes of the entire distribution of likely poses. This in turn allows for a more graceful treatment of geometric perception scenarios where establishing one-to-one matches between points is conceptually ill-defined, such as textureless, symmetrical and/or occluded objects and scenes where the correct pose is uncertain or there are multiple equally valid solutions.

Abstract: Developing embodied agents in simulation has been a key research topic in recent years. Exciting new tasks, algorithms, and benchmarks have been developed in various simulators. However, most of them assume deaf agents in silent environments, while we humans perceive the world with multiple senses. We introduce Sonicverse, a multisensory simulation platform with integrated audio-visual simulation for training household agents that can both see and hear. Sonicverse models realistic continuous audio rendering in 3D environments in real-time. Together with a new audio-visual VR interface that allows humans to interact with agents with audio, Sonicverse enables a series of embodied AI tasks that need audio-visual perception. For semantic audio-visual navigation in particular, we also propose a new multi-task learning model that achieves state-of-the-art performance. In addition, we demonstrate Sonicverse's realism via sim-to-real transfer, which has not been achieved by other simulators: an agent trained in Sonicverse can successfully perform audio-visual navigation in real-world environments. Sonicverse is available at: https://github.com/StanfordVL/Sonicverse.

Abstract: Three challenges limit the progress of robot learning research: robots are expensive (few labs can participate), everyone uses different robots (findings do not generalize across labs), and we lack internet-scale robotics data. We take on these challenges via a new benchmark: Train Offline, Test Online (TOTO). TOTO provides remote users with access to shared robotic hardware for evaluating methods on common tasks and an open-source dataset of these tasks for offline training. Its manipulation task suite requires challenging generalization to unseen objects, positions, and lighting. We present initial results on TOTO comparing five pretrained visual representations and four offline policy learning baselines, remotely contributed by five institutions. The real promise of TOTO, however, lies in the future: we release the benchmark for additional submissions from any user, enabling easy, direct comparison to several methods without the need to obtain hardware or collect data.

Abstract: We present Language-Image Value learning (LIV), a unified objective for vision-language representation and reward learning from action-free videos with text annotations. Exploiting a novel connection between dual reinforcement learning and mutual information contrastive learning, the LIV objective trains a multi-modal representation that implicitly encodes a universal value function for tasks specified as language or image goals. We use LIV to pre-train the first control-centric vision-language representation from large human video datasets such as EpicKitchen. Given only a language or image goal, the pre-trained LIV model can assign dense rewards to each frame in videos of unseen robots or humans attempting that task in unseen environments. Further, when some target domain-specific data is available, the same objective can be used to fine-tune and improve LIV and even other pre-trained representations for robotic control and reward specification in that domain. In our experiments on several simulated and real-world robot environments, LIV models consistently outperform the best prior input state representations for imitation learning, as well as reward specification methods for policy synthesis. Our results validate the advantages of joint vision-language representation and reward learning within the unified, compact LIV framework.