Autonomous robotic valve turning: A hierarchical learning approach

2014 IEEE International Conference on Robotics and Automation (ICRA)

Caldwell

2014

Self Cite

Abstract-This paper investigates application of machine learning to the problem of contact perception between a robot's gripper and an object. The input data comprises a multidimensional time-series produced by a force/torque sensor at the robot's wrist, the robot's proprioceptive information, namely, the position of the end-effector, as well as the robot's control command. These data are used to train a hidden Markov model (HMM) classifier. The output of the classifier is a prediction of the contact state, which includes no contact, a contact aligned with the central axis of the valve, and an edge contact. To distinguish between contact states, the robot performs exploratory behaviors that produce distinct patterns in the time-series data. The patterns are discovered by first analyzing the data using a probabilistic clustering algorithm that transforms the multidimensional data into a onedimensional sequence of symbols. The symbols produced by the clustering algorithm are used to train the HMM classifier. We examined two exploratory behaviors: a rotation around the xaxis, and a rotation around the y-axis of the gripper. We show that using these two exploratory behaviors we can successfully predict a contact state with an accuracy of 88 ± 5 % and 81 ± 10 %, respectively.

Section: Related Workmentioning

confidence: 99%

Robot-object contact perception using symbolic temporal pattern learning

Jamali

2014 IEEE International Conference on Robotics and Automation (ICRA)

Caldwell

2014

Self Cite

“…This paper proposes the use of Learning by Demonstration (LbD) to learn the trajectory using the knowledge of experimented Remotely Operated underwater Vehicles (ROV) pilots. This concept has been proved previously in a laboratory environment [3], [4]. The following properties of the LbD make it suitable for planning the movement of such intervention task: ease of representation and learning, compactness of the representation, robustness against perturbations and changes in a dynamic environment, ease of reuse for related tasks and easy modification for new tasks.…”

Section: Introductionmentioning

confidence: 98%

Towards valve turning with an AUV using Learning by Demonstration

Carrera

Carreras

2013 MTS/IEEE OCEANS - Bergen

et al. 2013

Self Cite

Abstract-AUVs have experimentally done the first steps to solve basic intervention tasks. First results have been promising in the task of retrieving an object from the seabed. In this paper we extend the complexity of the task with the help of a Learning by Demonstration (LbD) approach. An extension of a LbD algorithm to learn the pose and orientation of the trajectory is presented to achieve a valve turning task. A batch of demonstrations done in ROV mode is used by the LbD to learn the trajectory, taking advantage of the experience of the pilots. Moreover, the paper present a controller able to coordinate the movement of the manipulator using the position or the orientation of the end-effector and also moving the AUV when is required. Both systems have been tested together in a simulated environment to solve the task of interacting with a valve located on a ROV panel. The experiments has been done in an environment without perturbations and in an environment with different perturbations. The method has been able to overcome the perturbations and complete the task successfully. Furthermore, the proposed controller has simplified the use of the manipulator during the intervention task. The robot is equipped with a 4 DOFs manipulator having a griper as end effector to operate the T bar handles found in the panel. Panel and valve handle position and orientation are detected by a computer vision program based on template matching.

“…The RFDM system in our previous research [8], monitors the relative movement between the valve and the end-effector and generates decisions according to the defined linguistic rules. One of the drawbacks of the previous reactive system is that, it is independent of the distance between the gripper and the valve.…”

Section: Learning Of Reactive Behaviormentioning

confidence: 99%

“…The objective function can be minimized using various optimization algorithms. In [8], we applied four different optimization algorithms including gradient-descent, cross entropy method [19], covariance matrix adaptationevolution strategy (CMA-ES) [20], and modified Price algorithm [21]. The number of optimization parameters for this problem is equal to the number of membership functions multiplied by two (center and standard deviation), plus the number of constant outputs.…”

Section: B Tuning the Fuzzy Systemmentioning

confidence: 99%

“…Compared to our previous research [8], [9], this work provides the following three contributions: (i) in our previous research the turning phase was done by programming the turning motion into the robot. In this paper, on the other hand, a force control strategy is proposed for handling the turning phase; (ii) similar to the previous research a Reactive Fuzzy Decision Maker (RFDM) system is designed in order to react to external disturbances and sudden movements.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning reactive robot behavior for autonomous valve turning

Ahmadzadeh

2014 IEEE-RAS International Conference on Humanoid Robots

Jamisola

et al. 2014

Self Cite

Abstract-A learning approach is proposed for the challenging task of autonomous robotic valve turning in the presence of active disturbances and uncertainties. The valve turning task comprises two phases: reaching and turning. For the reaching phase the manipulator learns how to generate trajectories to reach or retract from the target. The learning is based on a set of trajectories demonstrated in advance by the operator. The turning phase is accomplished using a hybrid force/motion control strategy. Furthermore, a reactive decision making system is devised to react to the disturbances and uncertainties arising during the valve turning process. The reactive controller monitors the changes in force, movement of the arm with respect to the valve, and changes in the distance to the target. Observing the uncertainties, the reactive system modulates the valve turning task by changing the direction and rate of the movement. A real-world experiment with a robot manipulator mounted on a movable base is conducted to show the efficiency and validity of the proposed approach.