ICRA 2024 论文列表

This paper explores active sensing strategies that employ vision-based tactile sensors for robotic perception and classification of fabric textures. We formalize the active sampling problem in the context of tactile fabric recognition and provide an implementation of information-theoretic exploration strategies based on minimizing predictive entropy and variance of probabilistic models. Through ablation studies and human experiments, we investigate which components are crucial for quick and reliable texture recognition. Along with the active sampling strategies, we evaluate neural network architectures, representations of uncertainty, influence of data augmentation, and dataset variability. By evaluating our method on a previously published Active Clothing Perception Dataset and on a real robotic system, we establish that the choice of the active exploration strategy has only a minor influence on the recognition accuracy, whereas data augmentation and dropout rate play a significantly larger role. In a comparison study, while humans achieve 66.9% recognition accuracy, our best approach reaches 90.0% in under 5 touches, highlighting that vision-based tactile sensors are highly effective for fabric texture recognition.

We present a large empirical investigation on the use of pre-trained visual representations (PVRs) for training downstream policies that execute real-world tasks. Our study involves five different PVRs, each trained for five distinct manipulation or indoor navigation tasks. We performed this evaluation using three different robots and two different policy learning paradigms. From this e ort, we can arrive at three insights: 1) the performance trends of PVRs in the simulation are generally indicative of their trends in the real world, 2) the use of PVRs enables a first-of-its-kind result with indoor ImageNav (zero-shot transfer to a held-out scene in the real world), and 3) the benefits from variations in PVRs, primarily data-augmentation and fine-tuning, also transfer to the real-world performance. See project website1 for additional details and visuals.

Parking lots and autonomous warehouses for accommodating many vehicles/robots adopt designs in which the underlying graphs are well-connected to simplify planning and reduce congestion. In this study, we formulate and delve into the largest well-connected set (LWCS) problem and explore its applications in layout design for multi-robot path planning. Roughly speaking, a well-connected set over a connected graph is a set of vertices such that there is a path on the graph connecting any pair of vertices in the set without passing through any additional vertices of the set. Identifying an LWCS has many potential high-utility applications, e.g., for determining parking garage layout and capacity, as prioritized planning can be shown to be complete when start/goal configurations belong to an LWCS. In this work, we establish that computing an LWCS is NP-complete. We further develop optimal and near-optimal LWCS algorithms, with the near-optimal algorithm targeting large maps. A complete prioritized planning method is given for planning paths for multiple robots residing on an LWCS.

Offline reinforcement learning (RL) has been shown to be successfully modeled as a sequence modeling problem, drawing inspiration from the success of Transformers. Offline RL is often limited by the quality of the offline dataset, so offline-to-online RL is a more realistic setting. Online decision transformer (ODT) is an effective and representative sequence modeling-based offline-to-online RL method. Despite its effectiveness, ODT still suffers from the sample inefficiency problem during the online fine-tuning phase. This sample inefficiency problem arises because the agent treats all state-action pairs in the replay buffer equally when trying to learn from the replay buffer. In this paper, we propose a simple yet effective method, called weighting online decision transformer with episodic memory (WODTEM), to improve sample efficiency. We first attempt to introduce an episodic memory (EM) mechanism into the sequence modeling-based RL methods. By utilizing the EM mechanism, we propose a novel training objective with a weighting function, based on ODT, to improve sample efficiency. Experimental results on multiple tasks show that WODTEM can improve sample efficiency.

Optimal decision-making presents a significant challenge for autonomous systems operating in uncertain, stochastic and time-varying environments. Environmental variability over time can significantly impact the system’s optimal decision making strategy for mission completion. To model such environments, our work combines the previous notion of Time-Varying Markov Decision Processes (TVMDP) with partial observability and introduces Time-Varying Partially Observable Markov Decision Processes (TV-POMDP). We propose a twopronged approach to accurately estimate and plan within the TV-POMDP: 1) Memory Prioritized State Estimation (MPSE), which leverages weighted memory to provide more accurate time-varying transition estimates; and 2) an MPSE-integrated planning strategy that optimizes long-term rewards while accounting for temporal constraint. We validate the proposed framework and algorithms using simulations and hardware, with robots exploring a partially observable, time-varying environments. Our results demonstrate superior performance over standard methods, highlighting the framework’s effectiveness in stochastic, uncertain, time-varying domains.

Depth estimation models have shown promising performance on clear scenes but fail to generalize to adverse weather conditions due to illumination variations, weather particles, etc. In this paper, we propose WeatherDepth, a self-supervised robust depth estimation model with curriculum contrastive learning, to tackle performance degradation in complex weather conditions. Concretely, we first present a progressive curriculum learning scheme with three simple-to-complex curricula to gradually adapt the model from clear to relative adverse, and then to adverse weather scenes. It encourages the model to gradually grasp beneficial depth cues against the weather effect, yielding smoother and better domain adaption. Meanwhile, to prevent the model from forgetting previous curricula, we integrate contrastive learning into different curricula. By drawing reference knowledge from the previous course, our strategy establishes a depth consistency constraint between different courses toward robust depth estimation in diverse weather. Besides, to reduce manual intervention and better adapt to different models, we designed an adaptive curriculum scheduler to automatically search for the best timing for course switching. In the experiment, the proposed solution is proven to be easily incorporated into various architectures and demonstrates state-of-the-art (SoTA) performance on both synthetic and real weather datasets. Source code and data are available at https://github.com/wangjiyuan9/WeatherDepth.

The teleoperation of complex, kinematically redundant robots with loco-manipulation capabilities represents a challenge for human operators, who have to learn how to operate the many degrees of freedom of the robot to accomplish a desired task. In this context, developing an easy-to-learn and easy-to-use human-robot interface is paramount. Recent works introduced a novel teleoperation concept, which relies on a virtual physical interaction interface between the human operator and the remote robot equivalent to a "Marionette" control, but whose feedback was limited to only visual feedback on the human side. In this paper, we propose extending the "Marionette" interface by adding a wearable haptic interface to cope with the limitations given by the previous works. Leveraging the additional haptic feedback modality, the human operator gains full sensorimotor control over the robot, and the awareness about the robot’s response and interactions with the environment is greatly improved. We evaluated the proposed interface and the related teleoperation framework with naive users, assessing the teleoperation performance and the user experience with and without haptic feedback. The conducted experiments consisted in a loco-manipulation mission with the CENTAURO robot, a hybrid leg-wheel quadruped with a humanoid dual-arm upper body.

Obtaining three-dimensional information, especially the z-axis depth information, is crucial for robotic micromanipulation. Due to the unavailability of depth sensors such as lidars in micromanipulation setups, traditional depth acquisition methods such as depth from focus or depth from defocus directly infer depth from microscopic images and suffer from poor resolution. Alternatively, micromanipulation tasks obtain accurate depth information by detecting the contact between an end-effector and an object (e.g., a cell). Despite its high accuracy, only sparse depth data can be obtained due to its low efficiency. This paper aims to address the challenge of acquiring dense depth information during robotic cell micromanipulation. A weakly-supervised depth completion network is proposed to take cell images and sparse depth data obtained by contact detection as input to generate a dense depth map. A two-stage data augmentation method is proposed to augment the sparse depth data, and the depth map is optimized by a network refinement method. The experimental results show that the MAE value of the depth prediction error is less than 0.3 µm, which proves the accuracy and effectiveness of the method. This deep learning network pipeline can be seamlessly integrated with the robotic micromanipulation tasks to provide accurate depth information.

This paper tackles a localization problem in large-scale indoor environments with wayfinding maps. A wayfinding map abstractly portrays the environment, and humans can localize themselves based on the map. However, when it comes to using it for robot localization, large geometrical discrepancies between the wayfinding map and the real world make it hard to use conventional localization methods. Our objective is to estimate a robot pose within a wayfinding map, utilizing RGB images from perspective cameras. We introduce two different imagination modules which are inspired by how humans can comprehend and interpret their surroundings for localization purposes. These modules jointly learn how to effectively observe the first-person-view (FPV) world to interpret bird-eye-view (BEV) maps. Providing explicit guidance to the two imagination modules significantly improves the precision of the localization system. We demonstrate the effectiveness of the proposed approach using real-world datasets, which are collected from various large-scale crowded indoor environments. The experimental results show that, in 85% of scenarios, the proposed localization system can estimate its pose within 3m in large indoor spaces. Project Site: https://rllab-snu.github.io/projects/WayIL/

Accurate and robust navigation in unstructured environments requires fusing data from multiple sensors. Such fusion ensures that the robot is better aware of its surroundings, including areas of the environment that are not immediately visible but were visible at a different time. To solve this problem, we propose a method for traversability prediction in challenging outdoor environments using a sequence of RGB and depth images fused with pose estimations. Our method, termed WayFASTER (Waypoints-Free Autonomous System for Traversability with Enhanced Robustness), uses experience data recorded from a receding horizon estimator to train a self-supervised neural network for traversability prediction, eliminating the need for heuristics. Our experiments demonstrate that our method excels at avoiding obstacles, and correctly detects that traversable terrains, such as tall grass, can be navigable. By using a sequence of images, WayFASTER significantly enhances the robot’s awareness of its surroundings, enabling it to predict the traversability of terrains that are not immediately visible. This enhanced awareness contributes to better navigation performance in environments where such predictive capabilities are essential.

We propose WayEx, a new method for learning complex goal-conditioned robotics tasks from a single demonstration. Our approach distinguishes itself from existing imitation learning methods by demanding fewer expert examples and eliminating the need for information about the actions taken during the demonstration. This is accomplished by introducing a new reward function and employing a knowledge expansion technique. We demonstrate the effectiveness of WayEx, our waypoint exploration strategy, across six diverse tasks, showcasing its applicability in various environments. Notably, our method significantly reduces training time by ∼50% as compared to traditional reinforcement learning methods. WayEx obtains a higher reward than existing imitation learning methods given only a single demonstration. Furthermore, we demonstrate its success in tackling complex environments where standard approaches fall short. Appendix is available at: https://waypoint-ex.github.io.

Detecting air flows caused by phenomena such as heat convection is valuable in multiple scenarios, including leak identification and locating thermal updrafts for extending UAV flight duration. Unfortunately, the heat signature of these flows is often too subtle to be seen by a thermal camera. While convection also leads to fluctuations in air density and hence causes so-called schlieren – intensity and color variations in images – existing techniques such as Background-oriented schlieren (BOS) allow detecting them only against a known background and from a static camera, making these approaches unsuitable for moving vehicles. In this work we demonstrate the feasibility of visualizing air movement by predicting the corresponding schlieren-induced optical flow from a single greyscale image captured by a moving camera against an unfamiliar background. We first record and label a set of optical flows in an indoor setup using standard BOS techniques. We then train a convolutional neural network (CNN) by applying the previously collected optical flow distortions to a dataset containing a mixture of real and synthetically generated images to predict the two-dimensional optical flow from a single image. Finally, we evaluate our approach on the task of extracting the optical flow caused by schlieren from both a static and moving camera on previously unseen flow patterns and background images.

This study proposes a novel planning framework based on a model predictive control formulation that incorporates signal temporal logic (STL) specifications for task completion guarantees and robustness quantification. This marks the first-ever study to apply STL-guided trajectory optimization for bipedal locomotion push recovery, where the robot experiences unexpected disturbances. Existing recovery strategies often struggle with complex task logic reasoning and locomotion robustness evaluation, making them susceptible to failures due to inappropriate recovery strategies or insufficient robustness. To address this issue, the STL-guided framework generates optimal and safe recovery trajectories that simultaneously satisfy the task specification and maximize the locomotion robustness. Our framework outperforms a state-of-the-art locomotion controller in a high-fidelity dynamic simulation, especially in scenarios involving crossed-leg maneuvers. Furthermore, it demonstrates versatility in tasks such as locomotion on stepping stones, where the robot must select from a set of disjointed footholds to maneuver successfully.

Autonomous mobility tasks such as last-mile delivery require reasoning about operator-indicated preferences over terrains on which the robot should navigate to ensure both robot safety and mission success. However, coping with out of distribution data from novel terrains or appearance changes due to lighting variations remains a fundamental problem in visual terrain-adaptive navigation. Existing solutions either require labor-intensive manual data re-collection and labeling or use hand-coded reward functions that may not align with operator preferences. In this work, we posit that operator preferences for visually novel terrains, which the robot should adhere to, can often be extrapolated from established terrain preferences within the inertial-proprioceptive-tactile domain. Leveraging this insight, we introduce Preference extrApolation for Terrain-awarE Robot Navigation (PATERN), a novel framework for extrapolating operator terrain preferences for visual navigation. PATERN learns to map inertial-proprioceptive-tactile measurements from the robot’s observations to a representation space and performs nearest-neighbor search in this space to estimate operator preferences over novel terrains. Through physical robot experiments in outdoor environments, we assess PATERN’s capability to extrapolate preferences and generalize to novel terrains and challenging lighting conditions. Compared to baseline approaches, our findings indicate that PATERN 1 robustly generalizes to diverse terrains and varied lighting conditions, while navigating in a preference-aligned manner.

Widely adopted motion forecasting datasets sub-stitute the observed sensory inputs with higher-level abstractions such as 3D boxes and polylines. These sparse shapes are inferred through annotating the original scenes with perception systems’ predictions. Such intermediate representations tie the quality of the motion forecasting models to the performance of computer vision models. Moreover, the human-designed explicit interfaces between perception and motion forecasting typically pass only a subset of the semantic information present in the original sensory input. To study the effect of these modular approaches, design new paradigms that mitigate these limitations, and accelerate the development of end-to-end motion forecasting models, we augment the Waymo Open Motion Dataset (WOMD) with large-scale, high-quality, diverse LiDAR data for the motion forecasting task.The new augmented dataset (WOMD-LiDAR)1 consists of over 100,000 scenes that each spans 20 seconds, consisting of well-synchronized and calibrated high quality LiDAR point clouds captured across a range of urban and suburban geographies. Compared to Waymo Open Dataset (WOD), WOMDLiDAR dataset contains 100× more scenes. Furthermore, we integrate the LiDAR data into the motion forecasting model training and provide a strong baseline. Experiments show that the LiDAR data brings improvement in the motion forecasting task. We hope that WOMD-LiDAR will provide new opportunities for boosting end-to-end motion forecasting models.

In the field of domain adaptation (DA) on 3D object detection, most of the work is dedicated to unsupervised domain adaptation (UDA). Yet, without any target annotations, the performance gap between the UDA approaches and the fully-supervised approach is still noticeable, which is impractical for real-world applications. On the other hand, weakly-supervised domain adaptation (WDA) is an underexplored yet practical task that only requires few labeling effort on the target domain. To improve the DA performance in a cost-effective way, we propose a general weak labels guided self-training framework, WLST, designed for WDA on 3D object detection. By incorporating autolabeler, which can generate 3D pseudo labels from 2D bounding boxes, into the existing self-training pipeline, our method is able to generate more robust and consistent pseudo labels that would benefit the training process on the target domain. Extensive experiments demonstrate the effectiveness, robustness, and detector-agnosticism of our WLST framework. Notably, it outperforms previous state-of-the-art methods on all evaluation tasks. Code and models are available at https://github.com/jacky121298/WLST. Note that the complete version with appendix is available on arXiv.

Underwater vehicle manipulator systems (UVMS) are increasingly popular platforms for performing subsea operations that require precision manipulation. While there is high demand for fully autonomous or even semi-autonomous systems, most UVMS still require human support teams. Developing new hardware and algorithms for autonomous underwater manipulation is challenging. Simulations do not capture the full complexity of the underwater environment, and deploying a UVMS at sea for testing/validation is resource-intensive and expensive. In this paper, we present a physical testbed for underwater manipulation that bridges the gap between simulation and full field trials. The underWater Arm-Vehicle Emulator (WAVE) is a 10-degree of freedom system designed to replicate an inspection-class UVMS. WAVE includes an underwater perception sensor and has 2 operating modes: rigid or passive-mode. In passive-mode, the ROV body can pitch similar to how a dynamically-coupled underactuated UVMS without pitch control would rotate during manipulation tasks. To validate the overall design and passive pitch concept, we evaluated the testbed during underwater experiments in energetic conditions at a wave basin. To support continued research and development in underwater robotics, we make the design open-access and freely available to the community.

A robotic hand for social interaction should be capable of comfortable touch with humans. However, it is difficult to mount skin, tactile sensors, and driving mechanism required for human contact, especially holding hands, on a slender finger. In addition, in order to unitize the hand for easy use with any robot and maintainability, the mechanism must be contained within the small space of the fingers and palms. In this paper, we propose a human-sized five-fingered robotic hand named WARABI Hand. It is covered with multi-layored rubber skin to realize human-like soft and pleasant feel. Force sensors on each finger link detect contact with humans and adjust gripping force. We conducted experiments in which a humanoid equipped with WARABI Hand grasped forearm, held hands, and interlocked fingers with a person. The performance for object grasping was also evaluated. We demonstrated that our proposed hand is useful for interaction with humans including receiving and handing over things.