A Three-Layer Planning Architecture for the Autonomous Control of - - PowerPoint PPT Presentation

▶
a three layer planning architecture for the
SMART_READER_LITE
LIVE PREVIEW

A Three-Layer Planning Architecture for the Autonomous Control of - - PowerPoint PPT Presentation

Dec 13, 2022 210 likes •436 views

Journal Track A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Jos Carlos Gonzlez, Jos Carlos Pulido and Fernando Fernndez Planning and Learning Group Cognitive Systems

slide-1
SLIDE 1

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

José Carlos González, José Carlos Pulido and Fernando Fernández

29 June 2018 Computer Science Department

Planning and Learning Group

Journal Track Cognitive Systems Research (CSR),

  • vol. 43, pp. 232-249, Elsevier, June 2017,

doi:10.1016/j.cogsys.2016.09.003

slide-2
SLIDE 2

2/17

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

Introduction

Architecture

youtu.be/PbfqoILctH4

slide-3
SLIDE 3

3/17

Therapeutic procedure

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

Introduction

Architecture

slide-4
SLIDE 4

4/17

Automated Planning Levels

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Introduction

Architecture

High-level planning

High-level planning

Therapy configuration

Medium-level planning Low-level planning Humanoid robot

Actions

Kinect Sensor

Anthropometric data Low-level Instructions

Step A: Therapy definition Step B: Session execution

Planned sessions Perception Cognition Action

Therapy Designer Decision Support Robot Controller

slide-5
SLIDE 5

5/17

High-level planning – Therapy designer

  • Sessions have exercises

▪ Maximum and minimum duration ▪ Warming up, Training or Cooling down

  • Exercises have poses

▪ Set, perception, evaluation and correction

  • Variability and patient constraints

▪ Exercises cannot reappear in one session ▪ Exercise distribution should be assorted throughout sessions ▪ Avoid groups of exercises according to patient conditions

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Architecture

High-level planning

Medium-level planning

Exercises Sessions Poses Therapy

slide-6
SLIDE 6

6/17

High-level planning – Therapy designer

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Architecture

High-level planning

Medium-level planning

slide-7
SLIDE 7

7/17

  • TOCL (Therapeutic Objectives Cumulative Level)
  • Exercise attributes

▪ Adequacy level for each therapeutic objective ▪ Duration, intensity and difficulty ▪ Group of exercise (capability)

  • Problem goal: TOCLs reachability property

▪ The sum of the adequacy levels of the planned exercises must reach the respective TOCL for each session

Therapeutic Objectives TOCLs example

  • 1. Bimanual
  • 2. Fine unimanual
  • 3. Coarse unimanual
  • 4. Arm positioning
  • 5. Hand positioning

15 30 5

High-level planning – Therapy designer

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Architecture

High-level planning

Medium-level planning

slide-8
SLIDE 8

8/17

E3 E7 E5 E8 E4 E9

…

E2 E6

Exercise Database

E9 E1 E6 E1 E5 E8 E4 E7 E0 E5 E6 E1 E3 E8 E2 E0 S1 S2 S3

Planned sessions TOCLs Constraints

E1 E5 E9 E0

…

E3

TOCLs reachability property

High-level planning – Therapy designer

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Architecture

High-level planning

Medium-level planning

slide-9
SLIDE 9

9/17

High-level planning – Therapy designer

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Architecture

High-level planning

Medium-level planning

slide-10
SLIDE 10

10/17

Medium-level planning – Session control

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots High-level planning

Medium-level planning

Low-level planning

slide-11
SLIDE 11

11/17

Medium-level planning – Session control

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots High-level planning

Medium-level planning

Low-level planning

slide-12
SLIDE 12

12/17

Low-level planning - Independence

  • Generic low-level actions

▪ Can be interpreted by similar robots ▪ Only the Robot component has to be rewritten ▪ Tested with NAO, Ursus robot (UNEX) and REEM robot (PAL)

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Medium-level planning

Low-level planning

Conclusions

slide-13
SLIDE 13

13/17

Journal paper conclusions

  • NAOTherapist allows a humanoid robot to

autonomously drive therapeutic sessions previously planned by the system

  • The control is addressed at three abstraction levels

▪ High level: where the whole therapy is planned ▪ Medium level: where the session is controlled ▪ Low level: transparent for us, path-planning tasks

  • It has been evaluated with a large group of children…

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Low-level planning

Conclusions

Moving forward

slide-14
SLIDE 14

14/17

Moving forward - Evaluations

  • 2015 – Initial tests

▪ 120 healthy children in schools ▪ 3 real patients in a hospital

  • 2016 – Long-term tests

▪ 12 patients in a hospital ▪ 2 times per week for 4 months

  • 2017 – Intensive tests

▪ 25 patients in a summer camp ▪ Every day for 3 weeks

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Conclusions

Moving forward

References

  • First interaction tests
  • Children liked the system
  • Adjustements for real patients
  • Children were motivated
  • Our sessions were repetitive
  • Therapists were very interested
  • Polishing to develop a product
  • Several new activities
  • Therapists were very interested too
  • Children maintained their attention
slide-15
SLIDE 15

15/17

Moving forward - Improvements

  • High-level (therapy designer)

▪ Replanning for high-level events

  • Medium-level (session execution)

▪ Fully declarative mechanism for our decision support

‒ Planning domain ‒ Execution/monitoring of its actions ‒ Refinement of actions and abstraction of states

▪ Interruption of actions in the middle of their execution ▪ New games and interactive activities

‒ Simon with poses, storytelling...

  • Low-level

▪ Independence from the 3D sensor

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Conclusions

Moving forward

References

slide-16
SLIDE 16

16/17

Future work

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

n-level planning 1-level planning 0-level planning

0-actions

Scheduler-Planner

n-actions

. . .

Scheduler-Planner Scheduler-Planner

  • Development of a fully generic multilevel control architecture
  • Comparison with similar control systems

0-state n-state

. . .

Conclusions

Moving forward

References

slide-17
SLIDE 17

17/17

References

  • Enhancing a Robotic Rehabilitation Model for Hand-Arm Bimanual Intensive Therapy: Enrique García Estévez,

Irene Díaz Portales, José Carlos Pulido, Raquel Fuentetaja and Fernando Fernandez, on the 3rd Iberian Robotics Conference, (ROBOT 2017), Rehabilitation and Assistive Robotics special session, Seville (Spain), November 2017.

  • Evaluating the Child-Robot Interaction of the NAOTherapist Platform in Pediatric Rehabilitation: José Carlos

Pulido, José Carlos González, Cristina Suárez, Antonio Bandera, Pablo Bustos, Fernando Fernández. International Journal of Social Robotics, 2017.

  • A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social

Robots: José Carlos González, José Carlos Pulido, Fernando Fernández. Journal of Cognitive Systems Research, 2017.

  • Playing with Robots: An Interactive Simon Game: Mısra Turp, José Carlos Pulido, José Carlos González, Fernando

Fernández, in proceedings of the Workshop on Social Robotics and Human-Robot Interaction (RSIM), CAEPIA 2015 Albacete (Spain), 2015.

  • Therapy Monitoring and Patient Evaluation with Social Robots: Alejandro Martín, José Carlos González, José Carlos

Pulido, Ángel García-Olaya, Fernando Fernández and Cristina Suárez-Mejías, in proceedings of the 3rd Workshop on ICTs for improving Patients Rehabilitation Research Techniques, REHAB 2015 Lisbon (Portugal), 2015.

  • Planning, Execution and Monitoring of Physical Rehabilitation Therapies with a Robotic Architecture: José

Carlos González, José Carlos Pulido, Fernando Fernández and Cristina Suárez-Mejías, in proceedings of the 26th Medical Informatics Europe conference (MIE), Studies in Health Technology and Informatics, vol. 210, pp. 339-343, Madrid (Spain), 2015.

  • Goal-directed Generation of Exercise Sets for Upper-Limb Rehabilitation: José Carlos Pulido, José Carlos González,

Arturo González-Ferrer, Javier García, Fernando Fernández, Antonio Bandera, Pablo Bustos and Cristina Suárez, in proceedings of the 5th Workshop on Knowledge Engineering for Planning and Scheduling (KEPS), ICAPS conference, pp. 38-45, Portsmouth (New Hampshire, USA), 2014.

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Moving forward

References

slide-18
SLIDE 18

Planning and Learning Group

Thank you for your attention

29 June 2018 Computer Science Department

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

José Carlos González, José Carlos Pulido and Fernando Fernández

Cognitive Systems Research (CSR),

  • vol. 43, pp. 232-249, Elsevier, June 2017,

doi:10.1016/j.cogsys.2016.09.003 Journal Track

slide-19
SLIDE 19

19/17

Background

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots Cervical vertebrae

Brachial plexus

  • Cerebral palsy and brachial plexus palsy

▪ Children with upper-limb impairments ▪ They improve with rehabilitation

  • Very long and repetitive sessions

▪ Patients may lose interest ▪ A lot of time for clinicians

  • Autonomous social therapeutic robots

▪ Motivate patients ▪ Can capture clinical metrics on the fly

Spine

slide-20
SLIDE 20

20/17

Socially Assistive Robotic System

Patient

NAOTherapist

Social interaction Controls session Clinical metrics Clinical and technical support Therapist Technician and therapist

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

slide-21
SLIDE 21

21/17

NAOTherapist Architecture

▪ Modular ▪ Distributed ▪ Reusable

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots

slide-22
SLIDE 22

22/17

Medium-level planning – Session control

  • Replanning mechanism

▪ Mature planning framework ▪ Modular ▪ Check the actual effects in the environment ▪ Planner as a black box

  • Receives world states

▪ Transformed to full high-level states

  • Plans every deliberative low-level action

▪ Plans at high level ▪ High->Low-level action set

Robot High to Low Low to High Executive Monitoring Decision Support Low Actions Set Low State 1 2 3 4 5 6 7 8 9 10

PELEA planning framework

A Three-Layer Planning Architecture for the Autonomous Control of Rehabilitation Therapies Based on Social Robots