ICRA 2023 Papers

Robotic manipulation can benefit from wrist-mounted force/torque (F/T) sensors, but conventional F/T sensors can be expensive, difficult to install, and damaged by high loads. We present Visual Force/Torque Sensing (VFTS), a method that visually estimates the 6-axis F/T measurement that would be reported by a conventional F/T sensor. In contrast to approaches that sense loads using internal cameras placed behind soft exterior surfaces, our approach uses an external camera with a fisheye lens that observes a soft gripper. VFTS includes a deep learning model that takes a single RGB image as input and outputs a 6-axis F/T estimate. We trained the model with sensor data collected while teleoperating a robot (Stretch RE1 from Hello Robot Inc.) to perform manipulation tasks. VFTS outperformed F/T estimates based on motor currents, generalized to a novel home environment, and supported three autonomous tasks relevant to healthcare: grasping a blanket, pulling a blanket over a manikin, and cleaning a manikin's limbs. VFTS also performed well with a manually operated pneumatic gripper. Overall, our results suggest that an external camera observing a soft gripper can perform useful visual force/torque sensing for a variety of manipulation tasks.

This work presents a simple linear policy for direct force control for quadrupedal robot locomotion. The motivation is that force control is essential for highly dynamic and agile motions. We learn a linear policy to generate end-foot trajectory parameters and a centroidal wrench, which is then distributed among the legs based on the foot contact information using a quadratic program (QP) to get the desired ground reaction forces. Unlike the majority of the existing works that use complex nonlinear function approximators to represent the RL policy or model predictive control (MPC) methods with many optimization variables in the order of hundred, our controller uses a simple linear function approximator to represent policy along with only a twelve variable QP for the force distribution. A centroidal dynamics-based MPC method is used to generate reference trajectory data, and then the linear policy is trained using imitation learning to minimize the deviations from the reference trajectory. We demonstrate this compute-efficient controller on our robot Stoch3 in simulation and real-world experiments on indoor and outdoor terrains with push recovery.

Many surgical tasks require three or more tools working together, where a hands-free interface could extend a surgeon's actions to control a third surgical tool. However, most current interfaces do not allow skilled control of grasping critical to robotic manipulation. Here we first present a systematic study to identify efficient and intuitive interaction strategies to control grasping of a surgical tool. A series of experiments were conducted to evaluate six foot pressure-based gestures. Based on the results, three modular novel foot-machine interfaces were developed, which can be integrated with other motion control interfaces. The identified interaction strategies were implemented to control a laparoscopic tool in a surgical simulator, and evaluated in a user study. The results illustrate how naive participants can operate grasping yielding smooth and pick & place operation.

Recently, foot stepping strategies of humanoid robots have been actively developed for robust balancing of humanoids against disturbances. In this paper, a novel stepping algorithm adjusting double support phase (DSP) time is proposed. First, the stepping algorithm is proposed based on a model predictive control (MPC) framework for capture point (CP) control and footstep adjustment. Next, when the remaining step time is not enough to adjust the footstep, the DSP scaling method brings the next swing phase forward by reducing the DSP time, which enables the robot to maintain the balance robustly. The robust balance control performance of the proposed method is validated through simulations and experiments when the robot is walking in the presence of external pushes. A more stable balancing performance is realized compared to state-of-the-art stepping controllers.

Seamlessly integrating rules in Learning-from-Demonstrations (LfD) policies is a critical requirement to enable the real-world deployment of AI agents. Recently, Signal Temporal Logic (STL) has been shown to be an effective language for encoding rules as spatio-temporal constraints. This work uses Monte Carlo Tree Search (MCTS) as a means of integrating STL specification into a vanilla LfD policy to improve constraint satisfaction. We propose augmenting the MCTS heuristic with STL robustness values to bias the tree search towards branches with higher constraint satisfaction. While the domain-independent method can be applied to integrate STL rules online into any pre-trained LfD algorithm, we choose goal-conditioned Generative Adversarial Imitation Learning as the offline LfD policy. We apply the proposed method to the domain of planning trajectories for General Aviation aircraft around a non-towered airfield. Results using the simulator trained on real-world data showcase 60% improved performance over baseline LfD methods that do not use STL heuristics. [Code]11Codebase: https://github.com/castacks/mcts-stl-planning [Video]22Video: https://youtu.be/fiFCwc57MQs

Mobility, power, and price points often dictate that robots do not have sufficient computing power on board to run contemporary robot algorithms at desired rates. Cloud computing providers such as AWS, GCP, and Azure offer immense computing power and increasingly low latency on demand, but tapping into that power from a robot is non-trivial. We present FogROS2, an open-source platform to facilitate cloud and fog robotics that is included in the Robot Operating System 2 (ROS 2) distribution. FogROS2 is distinct from its predecessor FogROS1 in 9 ways, including lower latency, overhead, and startup times; improved usability, and additional automation, such as region and computer type selection. Additionally, FogROS2 gains performance, timing, and additional improvements associated with ROS 2. In common robot applications, FogROS2 reduces SLAM latency by 50 %, reduces grasp planning time from 14 s to 1.2 s, and speeds up motion planning 45x. When compared to FogROS1, FogROS2 reduces network utilization by up to 3.8x, improves startup time by 63 %, and network round-trip latency by 97 % for images using video compression. The source code, examples, and documentation for FogROS2 are available at https://github.com/BerkeleyAutomation/FogROS2, and is available through the official ROS 2 repository at https://index.ros.org/p/FogROS2/.

In order to efficiently learn a dynamics model for a task in a new environment, one can adapt a model learned in a similar source environment. However, existing adaptation methods can fail when the target dataset contains transitions where the dynamics are very different from the source environment. For example, the source environment dynamics could be of a rope manipulated in free space, whereas the target dynamics could involve collisions and deformation on obstacles. Our key insight is to improve data efficiency by focusing model adaptation on only the regions where the source and target dynamics are similar. In the rope example, adapting the free-space dynamics requires significantly less data than adapting the free-space dynamics while also learning collision dynamics. We propose a new method for adaptation that is effective in adapting to regions of similar dynamics. Additionally, we combine this adaptation method with prior work on planning with unreliable dynamics to make a method for data-efficient online adaptation, called FOCUS. We first demonstrate that the proposed adaptation method achieves statistically significantly lower prediction error in regions of similar dynamics on simulated rope manipulation and plant watering tasks. We then show on a bimanual rope manipulation task that FOCUS achieves data-efficient online learning, in simulation and in the real world.

There is extensive literature on perceiving road structures by fusing various sensor inputs such as lidar point clouds and camera images using deep neural nets. Leveraging the latest advance of neural architects (such as transformers) and bird-eye-view (BEV) representation, the road cognition accuracy keeps improving. However, how to cognize the “road” for automated vehicles where there is no well-defined “roads” remains an open problem. For example, how to find paths inside intersections without HD maps is hard since there is neither an explicit definition for “roads” nor explicit features such as lane markings. The idea of this paper comes from a proverb: it becomes a way when people walk on it. Although there are no “roads” from sensor readings, there are “roads” from tracks of other vehicles. In this paper, we propose FlowMap, a path generation framework for automated vehicles based on traffic flows. FlowMap is built by extending our previous work RoadMap [1], a light-weight semantic map, with an additional traffic flow layer. A path generation algorithm on traffic flow fields (TFFs) is proposed to generate human-like paths. The proposed framework is validated using real-world driving data and is amenable to generating paths for super complicated intersections without using HD maps.

Unmanned aerial vehicles (UAVs) are finding use in applications that place increasing emphasis on robustness to external disturbances including extreme wind. However, traditional multirotor UAV platforms do not directly sense wind; conventional flow sensors are too slow, insensitive, or bulky for widespread integration on UAVs. Instead, drones typically observe the effects of wind indirectly through accumulated errors in position or trajectory tracking. In this work, we integrate a novel flow sensor based on micro-electro-mechanical systems (MEMS) hot-wire technology developed in our prior work [1] onto a multirotor UAV for wind estimation. Our sensor is omnidirectional (in the plane), lightweight, fast, and accurate. In order to achieve superior hover performance in windy conditions, we train a ‘wind-aware’ residual-based controller via reinforcement learning using simulated wind gusts and their aerodynamic effects on the drone. In extensive hardware experiments, we demonstrate the wind-aware controller out-performing two strong ‘wind-unaware’ baseline controllers in challenging windy conditions. See: youtu.be/KWqkH9Z-338.

Modular self-assembling systems typically assume that modules are present to assemble. But in sparsely observed ocean environments modules of an aquatic modular robotic system may be separated by distances they do not have the energy to cross, and the information needed for optimal path planning is often unavailable. In this work we present a flow-based rendezvous and docking controller that allows aquatic robots in gyre-like environments to rendezvous with and dock to a target by leveraging environmental forces. This approach does not require complete knowledge of the flow, but suffices with imperfect knowledge of the flow's center and shape. We validate the performance of this control approach in both simulations and experiments relative to naive rendezvous and docking strategies and show that energy efficiency improves as the scale of the gyre increases.

Given a building floorplan, humans can localize themselves by matching the observation of the environment with the floorplan using geometric, semantic, and topological clues. Inspired by this insight, this paper proposes a learning- based topometric robot localization method FloorplanNet, which implements a match between a metric robot map and the potentially inaccurate building floorplan in nonuniform scales and different shapes by semantic information. The method uses a novel Graph Neural Network to learn descriptors of nodes from topometric graphs generated from the input maps. We demonstrate that our method can match the 3D point cloud sub-map generated by the robot during the SLAM process with the 2D map. Furthermore, we apply our map-matching algorithm for real-world robot localization. We evaluate our method on several publicly available real-world datasets. Even though our network is solely trained using simulation data, our method demonstrates high robustness and effectiveness in real- world indoor environments and outperforms the existing SOTA map-matching algorithms. We further develop a simulator that automatically creates and annotates the required training data to train our neural networks. The method and simulator are released at: https://github.com/fengdelin/FloorplanNet.git

This paper tackles the task of singulating and grasping paper-like deformable objects. We refer to such tasks as paper-flipping. In contrast to manipulating deformable objects that lack compression strength (such as shirts and ropes), minor variations in the physical properties of the paper-like deformable objects significantly impact the results, making manipulation highly challenging. Here, we present Flipbot, a novel solution for flipping paper-like deformable objects. Flipbot allows the robot to capture object physical properties by integrating exteroceptive and proprioceptive perceptions that are indispensable for manipulating deformable objects. Furthermore, by incorporating a proposed coarse-to-fine exploration process, the system is capable of learning the optimal control parameters for effective paper-flipping through proprioceptive and exteroceptive inputs. We deploy our method on a real-world robot with a soft gripper and learn in a self-supervised manner. The resulting policy demonstrates the effectiveness of Flipbot on paper-flipping tasks with various settings beyond the reach of prior studies, including but not limited to flipping pages throughout a book and emptying paper sheets in a box. The code is available here: https://robotll.github.io/Flipbot/.

Small uncrewed aerial systems (sUASs) are useful tools for 3D reconstruction due to their speed, ease of use, and ability to access high-utility viewpoints. Today, most aerial survey approaches generate a preplanned coverage pattern assuming a planar target region. However, this is inefficient since it results in superfluous overlap and suboptimal viewing angles and does not utilize the entire flight envelope. In this work, we propose active path planning for photogrammetric reconstruction. Our main contribution is a view utility function based on Fisher information approximating the offline reconstruction uncertainty. The metric enables online path planning to make in-flight decisions to collect geometrically informative image data in complex terrain. We evaluate our approach in a photorealistic simulation. A viewpoint selection study shows that our metric leads to faster and more precise reconstruction than state-of-the-art active planning metrics and adapts to different camera resolutions. Comparing our online planning approach to an ordinary fixed-wing aerial survey yields 3.2 × faster coverage of 16 ha undulated terrain without sacrificing precision.

In this paper, we address the problem of using visuo-tactile feedback for 6-DoF localization and 3D reconstruction of unknown in-hand objects. We propose FingerSLAM, a closed-loop factor graph-based pose estimator that combines local tactile sensing at finger-tip and global vision sensing from a wrist-mount camera. FingerSLAM is constructed with two constituent pose estimators: a multi-pass refined tactile-based pose estimator that captures movements from detailed local textures, and a single-pass vision-based pose estimator that predicts from a global view of the object. We also design a loop closure mechanism that actively matches current vision and tactile images to previously stored key-frames to reduce accumulated error. FingerSLAM incorporates the two sensing modalities of tactile and vision, as well as the loop closure mechanism with a factor graph-based optimization framework. Such a framework produces an optimized pose estimation solution that is more accurate than the standalone estimators. The estimated poses are then used to reconstruct the shape of the unknown object incrementally by stitching the local point clouds recovered from tactile images. We train our system on real-world data collected with 20 objects. We demonstrate reliable visuo-tactile pose estimation and shape reconstruction through quantitative and qualitative real-world evaluations on 6 objects that are unseen during training.

In robotic assisted surgeries, surgical tools are inserted into the human body via an incision point in the abdominal wall, which is imposed as a remote center of motion (RCM). The selection of the incision's point location in the human body is critical for the success of the surgical procedure. In this paper, we propose a simulation tool for finding the optimal incision point location, which can be utilized by the surgeon during the preoperative stage. The surgeon can plan the path/region of intervention as well as sensitive regions which should be protected from unintentional damage by the surgical tool on the preoperative images of internal organs. A target admittance model that enforces a candidate incision as a RCM is utilized in the simulation enhanced by a term for following the planned path. We propose a cost evaluation function taking into account metrics involving the distance of the tool from sensitive areas, the tool links maximum pressure on tumors and the robot's dexterity measure. The example of a tumor resection task is used with the simulation tool to demonstrate its use in finding the incision points that ensures minimal intraoperative risks and accurate task execution.

Exploration of unknown space with an autonomous mobile robot is a well-studied problem. In this work we broaden the scope of exploration, moving beyond the pure geometric goal of uncovering as much free space as possible. We believe that for many practical applications, exploration should be contextualised with semantic and object-level understanding of the environment for task-specific exploration. Here, we study the task of bothfinding specific objects in unknown space as well as reconstructing them to a target level of detail. We therefore extend our environment reconstruction to not only consist of a background map, but also object-level and semantically fused submaps. Importantly, we adapt our previous objective function of uncovering as much free space as possible in as little time as possible with two additional elements: first, we require a maximum observation distance of background surfaces to ensure target objects are not missed by image-based detectors because they are too small to be detected. Second, we require an even smaller maximum distance to the found objects in order to reconstruct them with the desired accuracy. We further created a Micro Aerial Vehicle (MAV) semantic exploration simulator based on Habitat in order to quantitatively demonstrate how our framework can be used to efficiently find specific objects as part of exploration. Finally, we showcase this capability can be deployed in real-world scenes involving our drone equipped with an Intel RealSense D455 RGB-D camera.

Traditional aerial vehicles have limitations in their capabilities due to actuator constraints, such as motor saturation. The hardware components and their arrangement are designed to satisfy specific requirements and are difficult to modify during operation. To address this problem, we introduce a versatile modular multi-rotor vehicle that can change its capabilities by reconfiguration. Our modular robot consists of homogeneous cuboid modules, propelled by quadrotors with tilted rotors. Depending on the number of modules and their configuration, the robot can expand its actuation capabilities. In this paper, we build a mathematical model for the actuation capability of a modular multi-rotor vehicle and develop methods to determine if a vehicle is capable of satisfying a task requirement. Based on this result, we find the optimal configurations for a given task. Our approach is validated in realistic \mathbf{3D}\mathbf{3D} simulations, showing that our modular system can adapt to tasks with varying requirements.

We introduce the Few-Shot Object Learning (FEWSOL) dataset for object recognition with a few images per object. We captured 336 real-world objects with 9 RGB-D images per object from different views. Fewsol has object segmentation masks, poses, and attributes. In addition, synthetic images generated using 330 3D object models are used to augment the dataset. We investigated (i) few-shot object classification and (ii) joint object segmentation and few-shot classification with state-of-the-art methods for few-shot learning and meta-learning using our dataset. The evaluation results show the presence of a large margin to be improved for few-shot object classification in robotic environments, and our dataset can be used to study and enhance few-shot object recognition for robot perception 11Dataset and code available at https://irvlutd.github.io/FewSOL.

In autonomous driving, the novel objects and lack of annotations challenge the traditional 3D LiDAR semantic segmentation based on deep learning. Few-shot learning is a feasible way to solve these issues. However, currently few-shot semantic segmentation methods focus on camera data, and most of them only predict the novel classes without considering the base classes. This setting cannot be directly applied to autonomous driving due to safety concerns. Thus, we propose a few-shot 3D LiDAR semantic segmentation method that predicts both novel and base classes simultaneously. Our method tries to solve the background ambiguity problem in generalized few-shot semantic segmentation. We first review the original cross-entropy and knowledge distillation losses, then propose a new loss function that incorporates the background information to achieve 3D LiDAR few-shot semantic segmentation. Extensive experiments on SemanticKITTI demonstrate the effectiveness of our method.