Novel Gaits for a Novel Novel Gaits for a Novel Crawling/Grasping - - PowerPoint PPT Presentation

▶

Oct 14, 2022 23 likes •300 views

Novel Gaits for a Novel Novel Gaits for a Novel Crawling/Grasping Mechanism Crawling/Grasping Mechanism Richard Voyles Department of Computer Science and Engineering University of Minnesota AMAM 2000 University of Minnesota Department of

SLIDE 1

University of Minnesota Department of Computer Science

Novel Gaits for a Novel Novel Gaits for a Novel Crawling/Grasping Mechanism Crawling/Grasping Mechanism

Richard Voyles

Department of Computer Science and Engineering University of Minnesota AMAM 2000

SLIDE 2

University of Minnesota Department of Computer Science

Outline Outline

Motivation

– DARPA Distributed Robotics Program – Rangers and Scouts

Design Details

– Limb Mechanism – Integral Force/Torque Sensors

Locomotion Gaits

– Swimming, Narrow, Wheel, Body-Roll

Gait Adaptation Strategy (future work)

SLIDE 3

University of Minnesota Department of Computer Science

Scouts and Rangers Scouts and Rangers

DARPA

Distributed Robotics Program

Hierarchical Heterogeneous

SLIDE 4

University of Minnesota Department of Computer Science

Ranger Ranger Macrobot Macrobot

Covers Long Distances Remote Brain for Scout Localization Capability Carries and Launches 10

Scouts

SLIDE 5

University of Minnesota Department of Computer Science

Scout Scout Microbot Microbot

Rolling/Hopping

Locomotion

Compass,

Tiltmeter, Communication

Camera /

Microphone / Vibration / Gas Sniffer

Reconnaissance

Surveillance Search/Rescue

SLIDE 6

University of Minnesota Department of Computer Science

Scout Design Constraints Scout Design Constraints

Form Factor: hard constraint on

body diameter (< 50mm)

Ruggedness: able to survive ballistic

launching

Adaptability: encounters a variety of

unknown conditions

SLIDE 7

University of Minnesota Department of Computer Science

Scout Limitations Scout Limitations

No Ability to Manipulate

(other than pushing)

Can’t Control Hop Height Can’t Locomote when

Headroom Only a Few Times Rubble Size

SLIDE 8

University of Minnesota Department of Computer Science

Low Headroom Scenario Low Headroom Scenario

Terrorism Warfare Earthquake Other

Natural Disasters

SLIDE 9

University of Minnesota Department of Computer Science

Alternate Design Goals Alternate Design Goals

Same Form Factor Same Ruggedness Same Power

Constraints

Greater Adaptability

(low headroom)

Dextrous Manipulation

Dual-Use Limbs

– Locomotion – Manipulation

“Conserve

Mechanism”

Locomote by

dragging the body

SLIDE 10

University of Minnesota Department of Computer Science

Outline Outline

Motivation

– DARPA Distributed Robotics Program – Rangers and Scouts

Design Details

– Limb Mechanism – Integral Force/Torque Sensors

Locomotion Gaits

– Swimming, Narrow, Wheel, Body-Roll

Gait Adaptation Strategy (future work)

SLIDE 11

University of Minnesota Department of Computer Science

TerminatorBot TerminatorBot -

Alternate Scout

Alternate Scout

Two 3-DoF Arms that Stow Inside Body Dual-Use Arms for both Locomotion

and Manipulation

Four Locomotion Gait Classes:

– “Swimming” Gaits (dry land) – Narrow Passage Gait (no wider than body) – “Bumpy Wheel” Rolling Gait – “Body-Roll” Dynamic Gait

SLIDE 12

University of Minnesota Department of Computer Science

TerminatorBot TerminatorBot Form Factor Form Factor

Stowed Configuration Deployed Configuration

Hemispherical side for smooth manipulation Concave claw for traction/digging

SLIDE 13

University of Minnesota Department of Computer Science

TerminatorBot TerminatorBot Design Design

1-Dof Elbow with Integral Force/Torque Sensor 2-DoF Shoulder Differential Drive Motors and Gearboxes (6 total)

SLIDE 14

University of Minnesota Department of Computer Science

Elbow Force/Torque Sensors Elbow Force/Torque Sensors

Maltese Cross is isotropic but

torque saturates

Top Figure has higher

sensitivity to forces of interest, but torque saturates

Bottom figure “equalizes”

torque

Use LVDTs to avoid cross-

coupling (Khatib, et al)

SLIDE 15

University of Minnesota Department of Computer Science

TerminatorBot TerminatorBot Prototype Prototype

1st Prototype is 75 mm in Diameter

(approx. 2x scale)

SLIDE 16

University of Minnesota Department of Computer Science

Outline Outline

Motivation

– DARPA Distributed Robotics Program – Rangers and Scouts

Design Details

– Limb Mechanism – Integral Force/Torque Sensors

Locomotion Gaits

– Swimming, Narrow, Wheel, Body-Roll

Gait Adaptation Strategy (future work)

SLIDE 17

University of Minnesota Department of Computer Science

Swimming Gait Video Swimming Gait Video

“Expected” Gait Variable Body Height Adaptable to Terrain

SLIDE 18

University of Minnesota Department of Computer Science

Narrow Narrow-

Passage Gait Video

Passage Gait Video

New Mechanisms

Often Suggest Novel Gaits (Yim, Xu, Pai)

Non-Controllable

Body Height

Good Traction Arms Stretched Out in

Front of Body

SLIDE 19

University of Minnesota Department of Computer Science

Bumpy Wheel Gait Video Bumpy Wheel Gait Video

Non-Controllable

Body Height

4 Coupled Motors Forearms “Roll”

Like the Spoke of a Wheel

SLIDE 20

University of Minnesota Department of Computer Science

Body Body-

Roll Gait

Roll Gait

Use Dynamics to Roll the Body Requires Smooth Surface Swing One Arm Across Body Use Other Arm as Reaction Force Tuck Swinging Arm to Roll Un-implemented at this time

SLIDE 21

University of Minnesota Department of Computer Science

One One-

Armed Gaits

Armed Gaits

Fail-safe Operation

– Emergency Homing Measures – Graceful Degradation of Mission Performance

Body-Roll Gaits

SLIDE 22

University of Minnesota Department of Computer Science

Outline Outline

Motivation

– DARPA Distributed Robotics Program – Rangers and Scouts

Design Details

– Limb Mechanism – Integral Force/Torque Sensors

Locomotion Gaits

– Swimming, Narrow, Wheel, Body-Roll

Gait Adaptation Strategy (future work)

SLIDE 23

University of Minnesota Department of Computer Science

Locomotion Primitives and Skills Locomotion Primitives and Skills

Patterned after prior work in

manipulation (Morrow, Voyles)

Gaits = Locomotion Skills

– Gaits composed of collections of primitives – Gait cycles through primitive space

SLIDE 24

University of Minnesota Department of Computer Science

Simple Primitives Simple Primitives

Open and Closed Loop

– Joint-space trajectory segment – World-space trajectory segment – Torque command – Guarded move (sensor response)

Evaluation Metrics Based on Port-Based Adaptable

Agent Architecture (PB3A)

SLIDE 25

University of Minnesota Department of Computer Science

Learn to Evaluate Gaits Learn to Evaluate Gaits

Metrics

– Vertical Visual Servoing Error – Kinesthetic Sense of Torque – Visual Odometry

Eigenspace-Based Learning Approach (PBD) Currently Applying Towards Conventional

Mobile Robots

Morph Gaits from Closed- to Open-Loop

SLIDE 26

University of Minnesota Department of Computer Science

Summary and Future Work Summary and Future Work

Robot Design Novel Gaits Topics Under Investigation

– Gait Adaptation to Terrain

Learning Progress Metrics
Gait Morphing

– Learning Manipulation Primitives

Primordial Adaptation (Eigenspace Method)
Programming by Demonstration

SLIDE 27

University of Minnesota Department of Computer Science

Sponsors Sponsors

DARPA / MTO

– Distributed Robotics Program – contract MDA972-98-C-0008

Air Force Research Lab