Tucker Hermans

Surgeon shortages have driven significant recent advances in robotic surgical assistance. To progress toward surgical automation, we address transoral central airway obstruction (CAO) removal, selecting this procedure for its relatively constrained environment and clear clinical objectives. Prior work has demonstrated successful CAO removal with single-port, minimally invasive robotic systems, making it a promising target for autonomy. We present a new autonomous system designed for CAO removal that integrates a learning-based tissue retraction policy with a model-based resection planner, deployed on a bimanual robotic platform with vision guidance. While our long-term motivation is minimally invasive intervention, we evaluate our approach in an open-surgery setting, enabling rapid development and testing. We evaluate our system in 12 ex vivo CAO removal procedures running at three distinct levels of autonomy. Our results indicate that the learned retraction policy performs similarly or better than a manually designed heuristic, with a mean surgeon-rated score (1-10) of 8.0± 1.6 compared to 6.8± 2.5, and a mean percent tumor volume removal of 113.9± 18.8% compared to 111.9± 13.3%. The model-based resections performed best when supervised, resulting in less tissue damage and better cutting accuracy. Furthermore, our approach produced clinically successful outcomes in all eight supervised autonomous trials. Notably, in one case, the system achieved complete tumor removal without any human intervention. Our overall findings underscore the importance of high-level supervision, revealing targets for future automation.

Constructing 3D representations of object geometry is critical for many robotics tasks, particularly manipulation problems. These representations must be built from potentially noisy partial observations. In this work, we focus on the problem of reconstructing a multi-object scene from a single RGBD image using a fixed camera. Traditional scene representation methods generally cannot infer the geometry of unobserved regions of the objects in the image. Attempts have been made to leverage deep learning to train on a dataset of known objects and representations, and then generalize to new observations. However, this can be brittle to noisy real-world observations and objects not contained in the dataset, and do not provide well-calibrated reconstruction confidences. We propose BRRP, a reconstruction method that leverages preexisting mesh datasets to build an informative prior during robust probabilistic reconstruction. We introduce the concept of a retrieval-augmented prior, where we retrieve relevant components of our prior distribution from a database of objects during inference. The resulting prior enables estimation of the geometry of occluded portions of the in-scene objects. Our method produces a distribution over object shape that can be used for reconstruction and measuring uncertainty. We evaluate our method in both simulated scenes and in the real world. We demonstrate the robustness of our method against deep learning-only approaches while being more accurate than a method without an informative prior. Through real-world experiments, we particularly highlight the capability of BRRP to enable successful dexterous manipulation in clutter.

Scene_synthesizer is a library for writing procedural scene generators in Python with a special focus on robot manipulation. The resulting scenes can be exported to various formats, enabling physics simulation or rendering data pipelines for training robotics or vision models.

This article extends recent work in magnetic manipulation of conductive, nonmagnetic objects using rotating magnetic dipole fields. Eddy-current-based manipulation provides a contact-free way to manipulate metallic objects. We are particularly motivated by the large amount of aluminum in space debris. We previously demonstrated dexterous manipulation of solid spheres with all object parameters known a priori. This work expands the previous model, which contained three discrete modes, to a continuous model that covers all possible relative positions of the manipulated spherical object with respect to the magnetic field source. We further leverage this new model to examine manipulation of spherical objects with unknown physical parameters by applying techniques from the online-optimization and adaptive-control literature. Our experimental results validate our new dynamics model, showing that we get improved performance compared to the previously proposed model, while also solving a simpler optimization problem for control. We further demonstrate the first physical magnetic manipulation of aluminum spheres, as previous controllers were only physically validated on copper spheres. We show that our adaptive control framework can quickly acquire useful object parameters when weakly initialized. Finally, we demonstrate that the spherical-object model can be used as an approximate model for adaptive control of nonspherical objects by performing magnetic manipulation of a variety of objects for which a spherical model is not an obvious approximation.

Eddy currents induced by rotating magnetic dipole fields can produce forces and torques that enable dexterous manipulation of conductive nonmagnetic objects. This paradigm shows promise for application in the remediation of space debris. The induced force from each rotating-magnetic-dipole field source always includes a repulsive component suggesting that the object should be surrounded by field sources to some degree to ensure the object does not leave the dexterous workspace during manipulation. In this paper, we show that it is possible to fully control the position of an object in a workspace near the midpoint between just two stationary field sources. A given position controller requires a low-level force controller. We propose two new force controllers, and compare them with the state-of-the-art method from the literature. One of the new force controllers is particularly good at not inducing parasitic torques, which is hypothesized to be beneficial for future tasks manipulating and de-tumbling rotating resident space objects. We perform experimental verification using numerical and physical simulators of microgravity.

Objects rarely sit in isolation in everyday human environments. If we want robots to operate and perform tasks in our human environments, they must understand how the objects they manipulate will interact with structural elements of the environment for all but the simplest of tasks. As such, we'd like our robots to reason about how multiple objects and environmental elements relate to one another and how those relations may change as the robot interacts with the world. We examine the problem of predicting inter-object and object-environment relations between previously unseen objects and novel environments purely from partial-view point clouds. Our approach enables robots to plan and execute sequences to complete multi-object manipulation tasks defined from logical relations. This removes the burden of providing explicit, continuous object states as goals to the robot. We explore several different neural network architectures for this task. We find the best performing model to be a novel transformer-based neural network that both predicts object-environment relations and learns a latent-space dynamics function. We achieve reliable sim-to-real transfer without any fine-tuning. Our experiments show that our model understands how changes in observed environmental geometry relate to semantic relations between objects. We show more videos on our website: https://sites.google.com/view/erelationaldynamics .

Robotic pick and place stands at the heart of autonomous manipulation. When conducted in cluttered or complex environments robots must jointly reason about the selected grasp and desired placement locations to ensure success. While several works have examined this joint pick-and-place problem, none have fully leveraged recent learning-based approaches for multi-fingered grasp planning. We present a modular algorithm for joint pick and place planning that can make use of state of the art grasp classifiers for planning multi-fingered grasps for novel objects from partial view point clouds. We demonstrate our joint pick and place formulation with several costs associated with different placement tasks. Experiments on pick and place tasks with cluttered scenes using a physical robot show that our joint inference method is more successful than a sequential pick then place approach, while also achieving better placement configurations.

Robotic pick and place stands at the heart of autonomous manipulation. When conducted in cluttered or complex environments robots must jointly reason about the selected grasp and desired placement locations to ensure success. While several works have examined this joint pick-and-place problem, none have fully leveraged recent learning-based approaches for multi-fingered grasp planning. We present a modular algorithm for joint pick and place planning that can make use of state of the art grasp classifiers for planning multi-fingered grasps for novel objects from partial view point clouds. We demonstrate our joint pick and place formulation with several costs associated with different placement tasks. Experiments on pick and place tasks with cluttered scenes using a physical robot show that our joint inference method is more successful than a sequential pick then place approach, while also achieving better placement configurations.

Objects rarely sit in isolation in everyday human environments. If we want robots to operate and perform tasks in our human environments, they must understand how the objects they manipulate will interact with structural elements of the environment for all but the simplest of tasks. As such, we'd like our robots to reason about how multiple objects and environmental elements relate to one another and how those relations may change as the robot interacts with the world. We examine the problem of predicting inter-object and object-environment relations between previously unseen objects and novel environments purely from partial-view point clouds. Our approach enables robots to plan and execute sequences to complete multi-object manipulation tasks defined from logical relations. This removes the burden of providing explicit, continuous object states as goals to the robot. We explore several different neural network architectures for this task. We find the best performing model to be a novel transformer-based neural network that both predicts object-environment relations and learns a latent-space dynamics function. We achieve reliable sim-to-real transfer without any fine-tuning. Our experiments show that our model understands how changes in observed environmental geometry relate to semantic relations between objects. We show more videos on our website: https://sites.google.com/view/erelationaldynamics.