Search for: All records

Tendon-driven continuum robots have been gaining popularity in medical applications due to their ability to curve around complex anatomical structures, potentially reducing the invasiveness of surgery. However, accurate modeling is required to plan and control the movements of these flexible robots. Physics-based models have limitations due to unmodeled effects, leading to mismatches between model prediction and actual robot shape. Recently proposed learning-based methods have been shown to overcome some of these limitations but do not account for hysteresis, a significant source of error for these robots. To overcome these challenges, we propose a novel deep decoder neural network that predicts the complete shape of tendon-driven robots using point clouds as the shape representation, conditioned on prior configurations to account for hysteresis. We evaluate our method on a physical tendon-driven robot and show that our network model accurately predicts the robot's shape, significantly outperforming a state-of-the-art physics-based model and a learning-based model that does not account for hysteresis. 

Lung cancer claims over 130,000 lives per year in the USA. For those with malignant tumors requiring resection, minimally invasive thoracic surgery via a video assisted or robotic approach is an alternative to highly invasive open thoracotomy (in which the chest is “cracked” open). This involves the insertion of 3-5 ports through the chest wall and the use of a camera and instruments mounted to rigid shafts, which are used to resect tissue in a deflated lung. One of these tools is typically a stapler which is able to simultaneously cut and seal the lung tissue. Tendon-driven continuum robots (TDCRs) are capable of curvilinear motions, which can add useful dexterity in constrained anatomical regions like the chest. However, the inherent flexibility of TDCRs presents challenges for integrating stapler-type end effectors. Lung staplers today are typically rigid tools because they require large axial forces to be transmitted along the tool shaft to fire staples. Such forces would apply large loads to curved continuum devices, changing their shapes and moving the end effector undesirably during staple firing. Low melting point alloys (LMPA) have been explored to stiffen substantially soft robots and compliant surgical devices. Here, we propose their use in a TDCR stapler to stiffen the tool shaft before staples are fired. Prior to stiffening, tendon actuation can provide enhanced maneuverability by curving the backbone compared to rigid staplers to position the stapler at the desired location. 

Automating robotic surgery via learning from demonstration (LfD) techniques is extremely challenging. This is because surgical tasks often involve sequential decisionmaking processes with complex interactions of physical objects and have low tolerance for mistakes. Prior works assume that all demonstrations are fully observable and optimal, which might not be practical in the real world. This paper introduces a sample-efficient method that learns a robust reward function from a limited amount of ranked suboptimal demonstrations consisting of partial-view point cloud observations. The method then learns a policy by optimizing the learned reward function using reinforcement learning (RL). We show that using a learned reward function to obtain a policy is more robust than pure imitation learning. We apply our approach on a physical surgical electrocautery task and demonstrate that our method can perform well even when the provided demonstrations are suboptimal and the observations are highdimensional point clouds. 

Tendon-driven continuum robot kinematic models are frequently computationally expensive, inaccurate due to unmodeled effects, or both. In particular, unmodeled effects produce uncertainties that arise during the robot’s operation that lead to variability in the resulting geometry. We propose a novel solution to these issues through the development of a Gaussian mixture kinematic model. We train a mixture density network to output a Gaussian mixture model representation of the robot geometry given the current tendon displacements. This model computes a probability distribution that is more representative of the true distribution of geometries at a given configuration than a model that outputs a single geometry, while also reducing the computation time. We demonstrate one use of this model through a trajectory optimization method that explicitly reasons about the 

Robotic surgical subtask automation has the potential to reduce the per-patient workload of human surgeons. There are a variety of surgical subtasks that require geometric information of subsurface anatomy, such as the location of tumors, which necessitates accurate and efficient surgical sensing. In this work, we propose an automated sensing method that maps 3D subsurface anatomy to provide such geometric knowledge. We model the anatomy via a Bayesian Hilbert map-based probabilistic 3D occupancy map. Using the 3D occupancy map, we plan sensing paths on the surface of the anatomy via a graph search algorithm, A * search, with a cost function that enables the trajectories generated to balance between exploration of unsensed regions and refining the existing probabilistic understanding. We demonstrate the performance of our proposed method by comparing it against 3 different methods in several anatomical environments including a real-life CT scan dataset. The experimental results show that our method efficiently detects relevant subsurface anatomy with shorter trajectories than the comparison methods, and the resulting occupancy map achieves high accuracy. 

Shape servoing, a robotic task dedicated to controlling objects to desired goal shapes, is a promising approach to deformable object manipulation. An issue arises, however, with the reliance on the specification of a goal shape. This goal has been obtained either by a laborious domain knowledge engineering process or by manually manipulating the object into the desired shape and capturing the goal shape at that specific moment, both of which are impractical in various robotic applications. In this paper, we solve this problem by developing a novel neural network DefGoalNet, which learns deformable object goal shapes directly from a small number of human demonstrations. We demonstrate our method’s effectiveness on various robotic tasks, both in simulation and on a physical robot. Notably, in the surgical retraction task, even when trained with as few as 10 demonstrations, our method achieves a median success percentage of nearly 90%. These results mark a substantial advancement in enabling shape servoing methods to bring deformable object manipulation closer to practical real-world applications. 

Soft robots have garnered great interest in recent years due to their ability to navigate complex environments and enhance safety during unplanned collisions. However, their softness typically limits the forces they can apply and payloads they can carry, compared to traditional rigid-link robots. In this paper we seek to create a hybrid manipulator that can switch between a state in which it acts as a soft robot, and a state in which it has a series of selectively stiffenable links. The latter state, accomplished by solidifying chambers of low melting point metal alloy within the robot, is in some ways analogous to a traditional rigid-link manipulator. It also has the added benefit that each “link” can be set to a desired straight or curved shape before solidification and re-shaped when desired. Thermoelectric heat pumps enable local heating and cooling of the alloy, and tendons running along the robot enable actuation. Using a simple two-link prototype, we illustrate how alloy melting and solidification can be used to modify the robot’s workspace and payload capacity. 

Conventional soft robots are designed with constant, passive stiffness properties, based on desired motion capabilities. The ability to encode two fundamentally different stiffness characteristics promises to enable a single robot to be optimized for multiple divergent tasks simultaneously and this has been previously proposed with a variety of approaches including jamming-based designs. In this paper, we propose phase-changing metallic spines of various geometries to independently control specific directional stiffness parameters of soft robots, changing how they respond to their actuation inputs and external loads. We fabricate spine-like structures using a low melting point alloy (LMPA), enabling us to switch on and off the effects of the stiff metal structure of the overall robot's stiffness during use. Changing soft robot morphology in this manner will enable these robots to adapt to environments and tasks that require divergent motion and force/moment application capabilities.