Abstract:Abstract-We describe the software components of a robotics system designed to autonomously grasp objects and perform dexterous manipulation tasks with only high-level supervision. The system is centered on the tight integration of several core functionalities, including perception, planning and control, with the logical structuring of tasks driven by a Behavior Tree architecture. The advantage of the implementation is to reduce the execution time while integrating advanced algorithms for autonomous manipulatio… Show more
“…Recent advances in motion planning, control, and perception have enabled robotic systems to perform complex manipulation tasks in real world domains [3], [8], [7], [31], [33]. These systems perform integrated task and motion planning (see e.g., [1], [6], [16], [40]) by using state machines or task graphs [4], [34] for high-level task specification and motion planning algorithms for realization of low-level subtasks.…”
Abstract-Robotic surgical assistants (RSAs) enable surgeons to perform delicate and precise minimally invasive surgery. Currently these devices are primarily controlled by surgeons in a local tele-operation (master-slave) mode. Introducing autonomy of surgical sub-tasks has the potential to assist surgeons, reduce fatigue, and facilitate supervised autonomy for remote tele-surgery. This paper considers the sub-task of surgical debridement: removing dead or damaged tissue fragments to allow the remaining healthy tissue to heal. We present an implemented automated surgical debridement system that uses the Raven, an open-architecture surgical robot with two cabledriven 7 DOF arms. Our system combines stereo vision for 3D perception, trajopt, an optimization-based motion planner, and model predictive control (MPC). Experiments with autonomous sensing, grasping, and removal of over 100 fragments suggest that it is possible for an autonomous surgical robot to achieve robustness comparable to human levels for a surgically-relevant subtask, although for our current implementation, execution time is 2-3× slower than human levels, primarily due to replanning times for MPC. This paper provides three contributions: (i) introducing debridement as a surgically-relevant sub-task for robotics, (ii) designing and implementing an autonomous multilateral surgical debridement system that uses both arms of the Raven surgical robot, and (iii) providing experimental data that highlights the importance of accurate state estimation for future research.
“…Recent advances in motion planning, control, and perception have enabled robotic systems to perform complex manipulation tasks in real world domains [3], [8], [7], [31], [33]. These systems perform integrated task and motion planning (see e.g., [1], [6], [16], [40]) by using state machines or task graphs [4], [34] for high-level task specification and motion planning algorithms for realization of low-level subtasks.…”
Abstract-Robotic surgical assistants (RSAs) enable surgeons to perform delicate and precise minimally invasive surgery. Currently these devices are primarily controlled by surgeons in a local tele-operation (master-slave) mode. Introducing autonomy of surgical sub-tasks has the potential to assist surgeons, reduce fatigue, and facilitate supervised autonomy for remote tele-surgery. This paper considers the sub-task of surgical debridement: removing dead or damaged tissue fragments to allow the remaining healthy tissue to heal. We present an implemented automated surgical debridement system that uses the Raven, an open-architecture surgical robot with two cabledriven 7 DOF arms. Our system combines stereo vision for 3D perception, trajopt, an optimization-based motion planner, and model predictive control (MPC). Experiments with autonomous sensing, grasping, and removal of over 100 fragments suggest that it is possible for an autonomous surgical robot to achieve robustness comparable to human levels for a surgically-relevant subtask, although for our current implementation, execution time is 2-3× slower than human levels, primarily due to replanning times for MPC. This paper provides three contributions: (i) introducing debridement as a surgically-relevant sub-task for robotics, (ii) designing and implementing an autonomous multilateral surgical debridement system that uses both arms of the Raven surgical robot, and (iii) providing experimental data that highlights the importance of accurate state estimation for future research.
“…To evaluate our algorithm, we conducted dozens of experiments with a robotic manipulation system [1]. In our experiments, a variety of unknown (manmade and natural) objects were placed on a table in front of the robot (e.g., Fig.…”
Section: Resultsmentioning
confidence: 99%
“…If an a priori CAD model of the target object is available, edge detection or primal sketches can be used to find a match [18]. When multiple images of an object are available, they can be used as templates to find the closet match [1]. The most important limitation of methods that rely on a priori models is that in unstructured environments such as our homes and offices, those models are unlikely to be available for all objects.…”
Section: B Object Detectionmentioning
confidence: 99%
“…To overcome this uncertainties, we rely on a library of compliant controllers which maintain proper contact with the environment during the robot's motion by responding to the detected contact forces. The robot motion is planned using CHOMP [1] to minimize contact with the environment. Our compliant controllers are described in detail in [15].…”
Section: Action Selection and Compliant Interactionmentioning
Abstract-We address the problem of clearing a pile of unknown objects using an autonomous interactive perception approach. Our robot hypothesizes the boundaries of objects in a pile of unknown objects (object segmentation) and verifies its hypotheses (object detection) using deliberate interactions. To guarantee the safety of the robot and the environment, we use compliant motion primitives for poking and grasping. Every verified segmentation hypothesis can be used to parameterize a compliant controller for manipulation or grasping. The robot alternates between poking actions to verify its segmentation and grasping actions to remove objects from the pile. We demonstrate our method with a robotic manipulator. We evaluate our approach with real-world experiments of clearing cluttered scenes composed of unknown objects.
“…Thus, our controllers must be robust to uncertainty in modeling and localization. Our system uses CHOMP [20] to plan a collision free trajectory to an action launch pose (i.e., robot hand pose) based on the COG and orientation of the facet. Then we execute a compliant controller which maintains proper contact with the environment by responding to the detected contact forces.…”
Abstract-Autonomous manipulation in unstructured environments presents roboticists with three fundamental challenges: object segmentation, action selection, and motion generation. These challenges become more pronounced when unknown manmade or natural objects are cluttered together in a pile. We present an end-to-end approach to the problem of manipulating unknown objects in a pile, with the objective of removing all objects from the pile and placing them into a bin. Our robot perceives the environment with an RGB-D sensor, segments the pile into objects using non-parametric surface models, computes the affordances of each object, and selects the best affordance and its associated action to execute. Then, our robot instantiates the proper compliant motion primitive to safely execute the desired action. For efficient and reliable action selection, we developed a framework for supervised learning of manipulation expertise. We conducted dozens of trials and report on several hours of experiments involving more than 1500 interactions. The results show that our learning-based approach for pile manipulation outperforms a common sense heuristic as well as a random strategy, and is on par with human action selection.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.