Updates on the NRI CSA grant

The National Robotics Initiative - Complementary Situational Awareness for Human-Robot Partnerships (NRI-CSA) project concluded recently:

"This was a 5-year collaborative project involving three collaborative teams at Vanderbilt, Carnegie Mellon, and Johns Hopkins University. The Principal investigators on this grant are Dr. Nabil Simaan (Vanderbilt), Dr. Howie Choset (CMU) and Dr. Russell H. Taylor (JHU).

The grant consists of three partner institutions contributing to laying the foundations to a new concept in robotics that we call Complementary Situational Awareness. Robots have been primarily used to augment human skill during manipulation tasks (e.g. for surgical applications, telemanipulation in hazardous environments) and in some cases to augment sensory presence (e.g. by providing force feedback to surgeons in cases where forces are below humanly perceptible thresholds). In our new approach robots will augment the human user not only in manipulation but also in understanding of the task and in action planning and execution. The idea is that the robots in some cases can sense things beyond human perception and this information may be used by the robot controller to create a model of the environment shape and the interaction characteristics. This robot situational awareness is then used to augment user/surgeon skill and situational awareness for carrying out complex tasks.

Key Project Highlights: Force-controlled Exploration for Updating Virtual Fixtures

This work deals with development of an approach for using exploration data to update and register an a-priori virtual fixture (high-level assistive telemanipulation law) geometry to a corresponding deformed and displaced physical geometry. This is representative of a robotic surgical intervention where a pre-operative surgical path-plan has to be updated due to organ intra-operative organ shift/swelling. Using hydrid force-motion control, exploration data (position and local surface normal) is used to deform and register the a-priori environment model to the exploration data set. The environment registration is achieved using a deformable registration approach based on coherent point drift. The task-description of the virtual fixture is then deformed and registered in the new environment and the new model is updated and used within a model-mediated telemanipulation framework. The approach is experimentally validated using a da-Vinci research kit (DVRK) master interface and a Cartesian stage robot.

Key Project Highlights: Smultaneous compliance and registration (SCAR) using stiffness and exploration data

Leveraging techniques pioneered by the SLAM community, we present a new filtering approach called simultaneous compliance and registration estimation or CARE. CARE is like SLAM in that it simultaneously determines the pose of a surgical robot while creating a map, but in this case, the map is a compliance map associated with a preoperative model of an organ as opposed to just positional information like landmark locations. The problem assumes that the robot is forcefully contacting and deforming the environment. This palpation has a dual purpose: 1) it provides the necessary geometric information to align or register the robot to a priori models, and 2) with palpation at varying forces, the stiffness/compliance of the environment can be computed. By allowing the robot to palpate its environment with varying forces, we create a force balanced spring model within a Kalman filter framework to estimate both tissue and robot position."

Source: NRI-CSA