This paper investigates how a novel paradigm called group-control can be effectively used for motion planning for microrobot swarms in a global field. We prove that Small-Time Local Controllability (STLC) in robot positions is achievable through group-control, with the minimum number of groups required for STLC being log_2(n + 2) + 1 for n robots. We then discuss the trade-offs between control complexity, motion planning complexity, and motion efficiency. We show how motion planning can be simplified if appropriate primitives can be achieved through more complex control actions. We identify motion planning problems that balance the number of motion primitives with planning complexity. Various instantiations of these motion planning problems are explored, with simulations used to demonstrate the effectiveness of group-control.

Autonomous medical robots hold promise to improve patient outcomes, reduce provider workload, democratize access to care, and enable superhuman precision. However, autonomous medical robotics has been limited by a fundamental data problem: existing medical robotic datasets are small, single-embodiment, and rarely shared openly, restricting the development of foundation models that the field needs to advance. We introduce Open-H-Embodiment, the largest open dataset of medical robotic video with synchronized kinematics to date, spanning more than 49 institutions and multiple robotic platforms including the CMR Versius, Intuitive Surgical's da Vinci, da Vinci Research Kit (dVRK), Rob Surgical BiTrack, Virtual Incision's MIRA, Moon Surgical Maestro, and a variety of custom systems, spanning surgical manipulation, robotic ultrasound, and endoscopy procedures. We demonstrate the research enabled by this dataset through two foundation models. GR00T-H is the first open foundation vision-language-action model for medical robotics, which is the only evaluated model to achieve full end-to-end task completion on a structured suturing benchmark (25

Few foundation models exist for robot-assisted surgery, partly because large robotic-surgery video corpora are difficult to assemble and existing models are evaluated mostly on laparoscopic benchmarks. Further, most existing models are evaluated on a small set of public benchmarks, mostly focused on laparoscopic surgery. We present OmniRAS, a family of 1B- and 2B-parameter V-JEPA-2.1 encoders for robot-assisted surgery, and detail their training. First, we release two densely annotated robotic-cholecystectomy datasets: OmniRAS-PR and a multi-label YT-Chole tool-verb-target task, the first triplet-style annotation for robotic cholecystectomy, together with splits, probe protocols, and an inter-rater study validating the shared phase ontology. Second, we document continued pretraining at up to 256 compute nodes with global batch 6,144 over 19 sources totaling approximately 2,650 hours of surgical video, 51

The KO (Kronecker Operation) code is a recent deep-learned error-correcting code using a neural network architecture to generalize a Reed-Muller code with Dumer decoding. Analyzing the encoder modules and using ablation techniques, we give interpretations of the KO encoder which significantly reduce the number of parameters. We also discuss interpretability aspects of the KO decoder. The interpretation opens up possibilities to give an explicit representation of KO codes, which could be useful for more efficient learning of KO codes and explaining the learning mechanism underlying the empirical observations made about its performance in previous work.

In recent years, the application of machine learning to minimally invasive surgery (MIS) has attracted considerable interest. Datasets are critical to the use of such techniques. This paper presents a unique dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers using the da Vinci Research Kit (dVRK). Unlike existing datasets, it addresses a critical gap by providing comprehensive kinematic data, recordings of all pedal inputs, and offers a time-stamped record of the endoscope's movements. This expanded version also includes segmentation and keypoint annotations of images, enhancing its utility for computer vision applications. Contributed by seven surgeons with varied backgrounds and experience levels that are provided as a part of this expanded version, the dataset is an important new resource for surgical robotics research. It enables the development of advanced methods for evaluating surgeon skills, tools for providing better context awareness, and automation of surgical tasks. Our work overcomes the limitations of incomplete recordings and imprecise kinematic data found in other datasets. To demonstrate the potential of the dataset for advancing automation in surgical robotics, we introduce two models that predict clutch usage and camera activation, a 3D scene reconstruction example, and the results from our keypoint and segmentation models.

Recent advances in large language models (LLMs) have demonstrated their potential as planners in human-robot collaboration (HRC) scenarios, offering a promising alternative to traditional planning methods. LLMs, which can generate structured plans by reasoning over natural language inputs, have the ability to generalize across diverse tasks and adapt to human instructions. This paper investigates the potential of LLMs to facilitate planning in the context of human-robot collaborative tasks, with a focus on their ability to reason from high-level, vague human inputs, and fine-tune plans based on real-time feedback. We propose a novel hybrid framework that combines LLMs with human feedback to create dynamic, context-aware task plans. Our work also highlights how a single, concise prompt can be used for a wide range of tasks and environments, overcoming the limitations of long, detailed structured prompts typically used in prior studies. By integrating user preferences into the planning loop, we ensure that the generated plans are not only effective but aligned with human intentions.

Robot assistants for older adults and people with disabilities need to interact with their users in collaborative tasks. The core component of these systems is an interaction manager whose job is to observe and assess the task, and infer the state of the human and their intent to choose the best course of action for the robot. Due to the sparseness of the data in this domain, the policy for such multi-modal systems is often crafted by hand; as the complexity of interactions grows this process is not scalable. In this paper, we propose a reinforcement learning (RL) approach to learn the robot policy. In contrast to the dialog systems, our agent is trained with a simulator developed by using human data and can deal with multiple modalities such as language and physical actions. We conducted a human study to evaluate the performance of the system in the interaction with a user. Our designed system shows promising preliminary results when it is used by a real user.

Robotic surgery offers enhanced precision and adaptability, paving the way for automation in surgical interventions. Cholecystectomy, the gallbladder removal, is particularly well-suited for automation due to its standardized procedural steps and distinct anatomical boundaries. A key challenge in automating this procedure is dissecting with accuracy and adaptability. This paper presents a vision-based autonomous robotic dissection architecture that integrates real-time segmentation, keypoint detection, grasping and stretching the gallbladder with the left arm, and dissecting with the other arm. We introduce an improved segmentation dataset based on videos of robotic cholecystectomy performed by various surgeons, incorporating a new “liver bed” class to enhance boundary tracking after multiple rounds of dissection. Our system employs state-of-the-art segmentation models and an adaptive boundary extraction method that maintains accuracy despite tissue deformations and visual variations. Moreover, we implemented an automated grasping and pulling strategy to optimize tissue tension before dissection upon our previous work. Ex vivo evaluations on porcine livers demonstrate that our framework significantly improves dissection precision and consistency, marking a step toward fully autonomous robotic cholecystectomy.

Several researchers have used deep learning to obtain novel feedback codes. Two such codes for AWGN channels with passive (possibly noisy) output feedback are Deepcode which employs a bit-by-bit rate 1/3 encoder, and Lightcode, which is a symbol-by-symbol code inspired by the Schalkwijk-Kailath (SK) scheme. Here, we build on prior work to interpret these codes by 1) providing the optimal maximum a posteriori (MAP) decoder for our simple non-linear interpretable encoder of a single-bit, two-round code that accurately approximates both single-bit Deepcode and Lightcode. This non-linear interpretable coding scheme, which mimics these codes, turns out to resemble both the functional form and performance of the Polyanskiy-Poor-Verdu (PPV) single bit feedback scheme that minimizes energy transmission asymptotically. 2) We extend our non-linear interpretable code to support more than one bit and two rounds, again providing an optimal MAP decoder. This remarkably simple and power-efficient nonlinear scheme provides insight into Lightcode.

Robotic surgery has reached a high level of maturity and has become an integral part of standard surgical care. However, existing surgeon consoles are bulky, take up valuable space in the operating room, make surgical team coordination challenging, and their proprietary nature makes it difficult to take advantage of recent technological advances, especially in virtual and augmented reality. One potential area for further improvement is the integration of modern sensory gloves into robotic platforms, allowing surgeons to control robotic arms intuitively with their hand movements. We propose one such system that combines an HTC Vive tracker, a Manus Meta Prime 3 XR sensory glove, and SCOPEYE wireless smart glasses. The system controls one arm of a da Vinci surgical robot. In addition to moving the arm, the surgeon can use fingers to control the end-effector of the surgical instrument. Hand gestures are used to implement clutching and similar functions. In particular, we introduce clutching of the instrument orientation, a functionality unavailable in the da Vinci system. The vibrotactile elements of the glove are used to provide feedback to the user when gesture commands are invoked. A qualitative and quantitative evaluation has been conducted that compares the current device with the dVRK console. The system is shown to have excellent tracking accuracy, and the new interface allows surgeons to perform common surgical training tasks with minimal practice efficiently.

In recent years, the potential applications of machine learning to Minimally Invasive Surgery (MIS) have spurred interest in data sets that can be used to develop data-driven tools. This paper introduces a novel dataset recorded during ex vivo pseudo-cholecystectomy procedures on pig livers, utilizing the da Vinci Research Kit (dVRK). Unlike current datasets, ours bridges a critical gap by offering not only full kinematic data but also capturing all pedal inputs used during the procedure and providing a time-stamped record of the endoscope's movements. Contributed by seven surgeons, this data set introduces a new dimension to surgical robotics research, allowing the creation of advanced models for automating console functionalities. Our work addresses the existing limitation of incomplete recordings and imprecise kinematic data, common in other datasets. By introducing two models, dedicated to predicting clutch usage and camera activation, we highlight the dataset's potential for advancing automation in surgical robotics. The comparison of methodologies and time windows provides insights into the models' boundaries and limitations.

Deep learning methods have recently been used to construct non-linear codes for the additive white Gaussian noise (AWGN) channel with feedback. However, there is limited understanding of how these black-box-like codes with many learned parameters use feedback. This study aims to uncover the fundamental principles underlying the first deep-learned feedback code, known as Deepcode, which is based on an RNN architecture. Our interpretable model based on Deepcode is built by analyzing the influence length of inputs and approximating the non-linear dynamics of the original black-box RNN encoder. Numerical experiments demonstrate that our interpretable model -- which includes both an encoder and a decoder -- achieves comparable performance to Deepcode while offering an interpretation of how it employs feedback for error correction.

Collaborative manipulation task often requires negotiation using explicit or implicit communication. An important example is determining where to move when the goal destination is not uniquely specified, and who should lead the motion. This work is motivated by the ability of humans to communicate the desired destination of motion through back-and-forth force exchanges. Inherent to these exchanges is also the ability to dynamically assign a role to each participant, either taking the initiative or deferring to the partner's lead. In this paper, we propose a hierarchical robot control framework that emulates human behavior in communicating a motion destination to a human collaborator and in responding to their actions. At the top level, the controller consists of a set of finite-state machines corresponding to different levels of commitment of the robot to its desired goal configuration. The control architecture is loosely based on the human strategy observed in the human-human experiments, and the key component is a real-time intent recognizer that helps the robot respond to human actions. We describe the details of the control framework, and feature engineering and training process of the intent recognition. The proposed controller was implemented on a UR10e robot (Universal Robots) and evaluated through human studies. The experiments show that the robot correctly recognizes and responds to human input, communicates its intent clearly, and resolves conflict. We report success rates and draw comparisons with human-human experiments to demonstrate the effectiveness of the approach.

We present an interpretation of Deepcode, a learned feedback code that showcases higher-order error correction relative to an earlier interpretable model. By interpretation, we mean succinct analytical encoder and decoder expressions (albeit with learned parameters) in which the role of feedback in achieving error correction is easy to understand. By higher-order, we mean that longer sequences of large noise values are acted upon by the encoder (which has access to these through the feedback) and used in error correction at the decoder in a two-stage decoding process.

Robotic surgery promises enhanced precision and adaptability over traditional surgical methods. It also offers the possibility of automating surgical interventions, resulting in reduced stress on the surgeon, better surgical outcomes, and lower costs. Cholecystectomy, the removal of the gallbladder, serves as an ideal model procedure for automation due to its distinct and well-contrasted anatomical features between the gallbladder and liver, along with standardized surgical maneuvers. Dissection is a frequently used subtask in cholecystectomy where the surgeon delivers the energy on the hook to detach the gallbladder from the liver. Hence, dissection along tissue boundaries is a good candidate for surgical automation. For the da Vinci surgical robot to perform the same procedure as a surgeon automatically, it needs to have the ability to (1) recognize and distinguish between the two different tissues (e.g. the liver and the gallbladder), (2) understand where the boundary between the two tissues is located in the 3D workspace, (3) locate the instrument tip relative to the boundary in the 3D space using visual feedback, and (4) move the instrument along the boundary. This paper presents a novel framework that addresses these challenges through AI-assisted image processing and vision-based robot control. We also present the ex-vivo evaluation of the automated procedure on chicken and pork liver specimens that demonstrates the effectiveness of the proposed framework.

Most deep-learned error-correcting codes (DL-ECCs) use binary cross-entropy (BCE) between the true input bits and the soft decoded outputs of the learned encoder/decoder pair as a loss function during training. It is known that this decomposes into two terms which in the case of a DL-ECC correspond to: a Kullback-Leibler (KL) divergence between the true and estimated posteriors on the input bits given the noisy channel outputs, and the conditional entropy (CE) of the input bits given the channel outputs. We use this decomposition to explore the training process of one particular DLECC termed TurboAE, which replaces constituent codes in the encoders/decoders of a Turbo code with learned convolutional neural networks. Evaluating each term in the BCE decomposition of TurboAE is facilitated through the junction tree algorithm for exact inference on graphs, yielding a MAP decoding of TurboAE-like codes of up to 40 input, 120 output bits (vs. the original 100-bit inputs). Plots of this decomposition over training offer insight into the alternating training process used in TurboAE and other DL-ECCs. We add to the growing body of work on interpreting DL-ECCs by providing a new lens through which to view any DL-ECC training process that uses BCE as a loss function.

Physical Human-Human Interaction (pHHI) involves the use of multiple sensory modalities. Studies of communication through spoken utterances and gestures are well established, but communication through force signals is not well understood. In this paper, we focus on investigating the mechanisms employed by humans during the negotiation through force signals, and how the robot can communicate task goals, comprehend human intent, and take the lead as needed. To achieve this, we formulate a task that requires active force communication and propose a taxonomy that extends existing literature. Also, we conducted a study to observe how humans behave during collaborative manipulation tasks. An important contribution of this work is the novel features based on force-kinematic signals that demonstrate predictive power to recognize symbolic human intent. Further, we show the feasibility of developing a real-time intent classifier based on the novel features and speculate the role it plays in high-level robot controllers for physical Human-Robot Interaction (pHRI). This work provides important steps to achieve more human-like fluid interaction in physical co-manipulation tasks that are applicable and not limited to humanoid, assistive robots, and human-in-the-loop automation.

As new deep-learned error-correcting codes continue to be introduced, it is important to develop tools to interpret the designed codes and understand the training process. Prior work focusing on the deep-learned TurboAE has both interpreted the learned encoders post-hoc by mapping these onto nearby "interpretable" encoders, and experimentally evaluated the performance of these interpretable encoders with various decoders. Here we look at developing tools for interpreting the training process for deep-learned error-correcting codes, focusing on: 1) using the Goldreich-Levin algorithm to quickly interpret the learned encoder; 2) using Fourier coefficients as a tool for understanding the training dynamics and the loss landscape; 3) reformulating the training loss, the binary cross entropy, by relating it to encoder and decoder parameters, and the bit error rate (BER); 4) using these insights to formulate and study a new training procedure. All tools are demonstrated on TurboAE, but are applicable to other deep-learned forward error correcting codes (without feedback).

This paper proposes a neural network-based user simulator that can provide a multimodal interactive environment for training Reinforcement Learning (RL) agents in collaborative tasks involving multiple modes of communication. The simulator is trained on the existing ELDERLY-AT-HOME corpus and accommodates multiple modalities such as language, pointing gestures, and haptic-ostensive actions. The paper also presents a novel multimodal data augmentation approach, which addresses the challenge of using a limited dataset due to the expensive and time-consuming nature of collecting human demonstrations. Overall, the study highlights the potential for using RL and multimodal user simulators in developing and improving domestic assistive robots.

Collaborative manipulation is inherently multimodal, with haptic communication playing a central role. When performed by humans, it involves back-and-forth force exchanges between the participants through which they resolve possible conflicts and determine their roles. Much of the existing work on collaborative human-robot manipulation assumes that the robot follows the human. But for a robot to match the performance of a human partner it needs to be able to take initiative and lead when appropriate. To achieve such human-like performance, the robot needs to have the ability to (1) determine the intent of the human, (2) clearly express its own intent, and (3) choose its actions so that the dyad reaches consensus. This work proposes a framework for recognizing human intent in collaborative manipulation tasks using force exchanges. Grounded in a dataset collected during a human study, we introduce a set of features that can be computed from the measured signals and report the results of a classifier trained on our collected human-human interaction data. Two metrics are used to evaluate the intent recognizer: overall accuracy and the ability to correctly identify transitions. The proposed recognizer shows robustness against the variations in the partner’s actions and the confounding effects due to the variability in grasp forces and dynamic effects of walking. The results demonstrate that the proposed recognizer is well-suited for implementation in a physical interaction control scheme.