Model-based autonomous system for performing dexterous, human-level manipulation tasks

Hudson, Nicolas; Ma, Jie; Jain, Abhinandan; Bajracharya, Max; Allen, Thomas F.; Sharan, Rangoli; Horowitz, Matanya B.; Kuo, Calvin; Howard, Thomas M.; Matthies, Larry; Backes, Paul; Burdick, Joel W.

doi:10.1007/s10514-013-9371-y

Cited by 15 publications

(17 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To this end we have used a fiducial detector from previous work (Hudson et al, 2014). By placing fiducials at known positions on the limbs, the 6-DOF fiducial pose in the robot frame (robot2f iducial) can be computed by forward kinematics.…”

Section: Perceptionmentioning

confidence: 99%

See 1 more Smart Citation

Mobile Manipulation and Mobility as Manipulation—Design and Algorithms of RoboSimian

Hebert

Bajracharya

et al. 2015

Journal of Field Robotics

Self Cite

116

View full text Add to dashboard Cite

This article presents the hardware design and software algorithms of RoboSimian, a statically stable quadrupedal robot capable of both dexterous manipulation and versatile mobility in difficult terrain. The robot has generalized limbs and hands capable of mobility and manipulation, along with almost fully hemispherical 3D sensing with passive stereo cameras. The system is semi-autonomous, enabling low-bandwidth, high latency control operated from a standard laptop. Because limbs are used for mobility and manipulation, a single unified mobile manipulation planner is used to generate autonomous behaviors, including walking, sitting, climbing, grasping, and manipulating. The remote operator interface is optimized to designate, parameterize, sequence, and preview behaviors, which are then executed by the robot. RoboSimian placed fifth in the DARPA Robotics Challenge (DRC) Trials, demonstrating its ability to perform disaster recovery tasks in degraded human environments.

show abstract

Section: Perceptionmentioning

confidence: 99%

“…The open-chain planner is simply an RRT-Connect planner (Kuffner & LaValle, 2000) with greedy path smoothing (Hudson et al, 2014). It can be used on any number of degrees-of-freedom of a serial manipulator, but we typically only use it for a single 7-DOF limb, attached at the chassis shoulder.…”

Section: Mobile Manipulation Motion Planningmentioning

confidence: 99%

Mobile Manipulation and Mobility as Manipulation—Design and Algorithms of RoboSimian

Hebert

Bajracharya

et al. 2015

Journal of Field Robotics

Self Cite

116

View full text Add to dashboard Cite

show abstract

“…Each IKfast call for the limb or torso requires fixing one joint, and solves for the IK of the remaining 6 joints in the limb or torso (resulting in up to 8 configurations). Each IKfast call for a Surrogate limb takes approximately 1000 microseconds, in contrast to a Barrett limb (with a wrist) which takes approximately 5 microseconds [23]. As the redundant space in the main serial chain is so high (8 DOF), searching over this space and using analytic IK to solve for joint angles is intractable.…”

Section: Implementation With Surrogatementioning

confidence: 99%

A Quadratic Programming Approach to Quasi-Static Whole-Body Manipulation

Shankar

Burdick

Hudson

2015

Springer Tracts in Advanced Robotics

View full text Add to dashboard Cite

Abstract. This paper introduces a local motion planning method for robotic systems with manipulating limbs, moving bases (legged or wheeled), and stance stability constraints arising from the presence of gravity. We formulate the problem of selecting local motions as a linearly constrained quadratic program (QP), that can be solved efficiently. The solution to this QP is a tuple of locally optimal joint velocities. By using these velocities to step towards a goal, both a path and an inverse-kinematic solution to the goal are obtained. This formulation can be used directly for real-time control, or as a local motion planner to connect waypoints. This method is particularly useful for high-degree-of-freedom mobile robotic systems, as the QP solution scales well with the number of joints. We also show how a number of practically important geometric constraints (collision avoidance, mechanism self-collision avoidance, gaze direction, etc) can be readily incorporated into either the constraint or objective parts of the formulation. Additionally, motion of the base, a particular joint, or a particular link can be encouraged/discouraged as desired. We summarize the important kinematic variables of the formulation, including the stance Jacobian, the reach Jacobian, and a center of mass Jacobian. The method is easily extended to provide sparse solutions, where the fewest number of joints are moved, by iteration using Tibshirani's method to accommodate an l1 regularizer. The approach is validated and demonstrated on SURROGATE, a mobile robot with a TALON base, a 7 DOF serial-revolute torso, and two 7 DOF modular arms developed at JPL/Caltech.

show abstract

“…Other works add features to user interfaces to reduce operator workload [13]. Some groups have attempted to make dexterous manipulators fully autonomous, for example, the DARPA ARM project [11]. Results generally point to fragility in the solution due the complexity of the task.…”

Section: Introductionmentioning

confidence: 99%

Supervisory control of a humanoid robot in microgravity for manipulation tasks

Farrell

Strawser

Hambuchen

et al. 2017

2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Abstract-Teleoperation is the dominant form of dexterous robotic tasks in the field. However, there are many use cases in which direct teleoperation is not feasible such as disaster areas with poor communication as posed in the DARPA Robotics Challenge, or robot operations on spacecraft a large distance from Earth with long communication delays. Presented is a solution that combines the Affordance Template Framework for object interaction with TaskForce for supervisory control in order to accomplish high level task objectives with basic autonomous behavior from the robot. TaskForce, is a new commanding infrastructure that allows for optimal development of task execution, clear feedback to the user to aid in off-nominal situations, and the capability to add autonomous verification and corrective actions. This framework has allowed the robot to take corrective actions before requesting assistance from the user. This framework is demonstrated with Robonaut 2 removing a Cargo Transfer Bag from a simulated logistics resupply vehicle for spaceflight using a single operator command. This was executed with 80% success with no human involvement, and 95% success with limited human interaction. This technology sets the stage to do any number of high level tasks using a similar framework, allowing the robot to accomplish tasks with minimal to no human interaction.

show abstract

Model-based autonomous system for performing dexterous, human-level manipulation tasks

Cited by 15 publications

References 16 publications

Mobile Manipulation and Mobility as Manipulation—Design and Algorithms of RoboSimian

Mobile Manipulation and Mobility as Manipulation—Design and Algorithms of RoboSimian

A Quadratic Programming Approach to Quasi-Static Whole-Body Manipulation

Supervisory control of a humanoid robot in microgravity for manipulation tasks

Contact Info

Product

Resources

About