Domo Partner Robots for Manipulation Aaron Edsinger MIT

Domo: Partner Robots for Manipulation Aaron Edsinger MIT Computer Science and Artificial Intelligence Laboratory Humanoid Robotics Group edsinger@csail. mit. edu 1. 26. 2006

Robots That Can Work Alongside Humans • Built for human environments • Safety in the human workspace • Humanoid body to work with everyday objects • Perform tasks that are important to people using natural strategies with everyday objects

Robots That Can Work Alongside Humans • Human-robot interface: use the body, not the keyboard. • Apply human expertise: coach the robot, through physical, visual, and auditory guidance.

Robots That Can Work Alongside Humans

Confronting Unstructured Environments

Robust Manipulation and Human Interactions in Unstructured Environments • Active Perception: Use the robot body to assist perception • Human Safety: Passive compliance and force control • Integration: Architecture to support rich, tightly coupled behaviors • Perceptual constraints: Focus resources on only taskrelevant features of objects and interactions

Domo • 29 active degrees of freedom (DOF) • Two 6 DOF force controlled arms using Series Elastic Actuators • Two 6 DOF force controlled hands using SEAs • A 2 DOF force controlled neck using SEAs • Stereo pair of Point Grey Firewire CCD cameras • Stereo Videre STH-DCSG-VAR-C Firewire cameras • Intersense 3 axis gyroscope • Two 4 DOF hands using Force Sensing Compliant (FSC) actuators

Domo • Embedded brushless and brushed DC motor drivers • 5 Embedded Motorola 56 F 807 DSPs running a 1 khz control loop • 4 CANBus channels providing 100 hz communication to external computation. • 49 potentiometers, 7 encoders, 24 tactile sensors, 12 brushless amplifiers, 17 brushed amplifiers, 12 sensor conditioners embedded on -board • An estimated 500 fabricated mechanical components and 60 electronics PCBs • 15 node Debian Linux cluster running a mixture of C/C++/Python and utilizing the Yarp and pysense robot libraries.

Series Elastic and Force Sensing Compliant Actuators F=-kx

Series Elastic and Force Sensing Compliant Actuators • Mechanically simple • Improved stability • Shock tolerance • Highly backdrivable • Low-grade components • Low impedance at high frequencies

Passive and Active Compliance Series Elastic Actuator Force based grasping

Domo: Behavior Based Architecture • Architectural primitives allow tightly integrated system • 100 hz scheduler • Dynamic arbitration • ~15 node Linux cluster • ~50 threads currently Homeostat

Architecture Example Arm Behaviors Head Behaviors

Behavior Based Architecture Arm Behaviors Head Behaviors

Detection of Human Interaction Forces Interaction forces at hands are approximately equal and opposite Interaction forces present Interaction forces not present

Detection of Human Interaction Forces Ballistic reaching: prediction error Efference copy model generates torque prediction. Torque prediction errors drive visual attention system. External forces: prediction error

Learning Tool Use By Demonstration • Motion feature points for tip detection • 3 D position estimation using probabilistic model

Estimation of Tool Position in the Hand

Autonomous Detection and Control of Human Tools by Demonstration