Thank  you.  I'm  going to  ask  you  all  to  turn  off  your  phones  if  you  can. And  also,  please  do  remember to  touch  your  trash  before  leaving. I'll  try  and  d  you  later  on. We'll  keep  the  nice  new  floors and  everything  looking  great. So  I'm  going  to  introduce  today's  speaker. Doctor  Kyle  Yasha. He's  currently  a  peral  fellow  at Wash  U  and  is  an  incoming  assistant  at  UCLA. Mechanical  Engineering,  school  of mechanical  engineering  and  aerospace, but  he's  more  on  the  mechanical  side. So  don't  ask  about  foes. Yeah.  Ch  you  want.  He  received the  NSF  graduate  research  fellowship, and  also  the  Washington  Research Foundation  Postdoctoral  Research  Fellowship, does  a  lot  of  really  interesting  work around  soft  robotics, Haptics,  perception,  human  robot  interfaces. And  a  lot  of  it  comes  from  his  background  our  shared, maybe  heritage  as  Native  wine  folks,  Kanaka. And  he's  going  to  talk  a  little  bit  about  that. Something  really  cool  that  I  want  to  highlight, and  part  of  the  reason  I  know, Kyle  was  he's  the  executive  director  of  Hona  Scholars. So  that  is  a  STEM  mentorship  program, recognized  as  one  of  the  top  ten  native  STEM  enterprises by  the  American  Indian  Science  and  Engineering  Society. He  has  done  amazing  work  in  that  area, that  kind  of  goes  far  above  and  beyond  how you  support  your  community. So  heading  that  off, I'll  headed  over  to  you,  Kyle.  Let's  welcome  our  speaker. Thanks. All  right.  Thank  you  for the  intro.  We're  good  on  the  mic. I  hear  myself.  Yeah,  thank  you  guys  all  for  coming. The  lunch  is  very  amazing. Today,  I'm  going  to  talk  about  some  of  my  work  in Haptics  and  soft  robotics  and also  a  flavor  of  kind  of that  community  integration  that  was  just  mentioned. I'll  start  off  a  little  bit  about  me. I  was  born  and  raised  Oahu. I  went  to  oma  High  School. Yep.  It's  where  all  the  Novan  go. There's  also  Puna  Ho  where  Obama  went, but  that's  our  Enemy  school  that  we  don't  talk  about. Some  fun  things  I  like  to  do  are  scuba  diving. This  is  one  of  the  shipwrecks  in  Waikiki. There's  a  few  that  you  can  get  to. Then  I  also  really  like  hiking. Particularly  hikes  through  forests. I  am  scared  of  heights. After  graduating  from  high  school, I  went  on  to  Harvard  where  I  did  my bachelor's  in  bio  engineering, and  then  I  also  did  a  minor  in  African  studies. That  also  helps  shape a  lot  of  the  viewpoints  that  I  have  in  engineering. Stanford  or  I  did  my  PhD  and masters  with  doctor  Allison  Ocamaro, at  the  collaborative  Haptics and  robotics  and  Medicine  Lab. And  currently,  I'm  a  post  fellow  at  Washington  State, mainly  working  on  kind  of  soft  robots  for  orchards, like  apple  picking,  and  also  interaction  with  humans. And  then,  as  mentioned  earlier, I'll  be  starting  as  an  assistant  professor  at CLA  next  this  coming  July  July  2025. Oh.  Thank  you. There's  kind  of  three  main  pillars  of  my  research  in developing  new  haptics  and robotic  systems  to  improve  daily  life. The  three  pillars  are  tics  soft  robotics, and  also  community  engagement. So  in  the  realm  of  haptic, some  things  that  I  work  on are  developing  new  haptic  devices, mainly  wearable  haptic  devices, and  understanding  how  people  perceive  these  devices. So  conducting  user  studies  around  that. In  the  realm  of  outreach  community  engagement, I  work  on  outreach  and  education. Some  projects  that  I'm  currently  working  on  are, how  can  we  use  use native  values  to  inspire  the  engineering  design  process, using  culturally  based  ways of  knowledge  and  ways  of  thinking and  how  we  actually contribute  to  our  work  in  engineering  and  science. In  the  almost  robotics, some  work  on  actuator  design and  trying  to  build new  actuators  that  do  different  types  of  things. Then  at  the  intersection  of  these  three  bubbles. I  also  have  a  few  more  collaborative  projects. So  one  that  I'll  talk  about  today  are  these  crowd sourced  and  clinical  perception tests  that  use  smartphones. So  the  idea  that  we  can  do  tic  studies, but  also  integrate  the  broader  community  by having  them  participate  in  our  experiments  more broadly  and  use  it  as  an  educational  tool  to  people  who might  not  have  as  much  exposure to  what  we  do  inside  the  lab. I  also  work  on  some  collaborative  robots, robots  for  agriculture, for  environmental  monitoring  and  surveying. This  is  a  project  that  we're currently  working  on  with  apple  picking  in  Washington, working  closely  with  farmers and  people  who  run  orchards  and  seeing  how  we can  develop  really  low  cost  soft  robots to  work  alongside  current  farm  workers. Also  at  the  intersection  of  Haptics  and  soft  robotics, there's  a  lot  of  new  ways  that  we  can  build human  robot  interfaces  to  control  these  robots, such  as  using  wearable  displays  and  new  tracking  systems. Then  also  at  the  center, there's  this  area  of  social  robot  interfaces where  we  can  use  tools  in  Haptics  and soft  robotics  that  influence  things  like  healthcare, different  effective  feedback  methods. For  today,  I'll  mainly  focus  on  these  areas looking  at  some  of my  lab  work  in  Haptics  and  soft  robotics, and  then  how  I  use  some  of  the  tools  from my  background  to  bring  it outside  of  the  lab  and  into  the  community. I'll  talk  about  a  wearable  Haptic  device, the  massive  open  online  platform  for  Haptics, which  uses  these  cell  phones,  fiber  reinforce, soft  actuators,  and  some  of the  current  and  future  work  on  collaborative  robotics. First,  I'll  talk  about  this  three  degree  of freedom  pneumatic  Haptic  device that  was  created  for  the  forearm. This  is  presented  in  two  paper  is a  conference  paper  and  a  transactions  on  Haptics  paper. But  before  getting  into the  nitty  gritty  details  of  the  work, I'll  preface  this  with  some  background  in  Haptics because  it  might  be  not  as  familiar  to  everyone. Haptics  or  relating  to  the  sense  of  touch. So  Haptic  feedback  comes  in  two  main  pathways. The  first  is  kinesthetic. And  kinesthetic  deals  with  forces  that are  often  across  or  across  joints. So  we  can  think  of  kinesthetic  feedback as  the  location  and  configuration  of  our  limbs  in  space. Kind  of  the  motion  of  those  limbs, the  force  large  forces  that  we  feel  like carrying  a  heavy  bottle  while  we're  pouring  water, and  also  like  compliance  of  different  systems that  we  feel  kind  of  like  through  our  joints. Kind  of  like  trying  to  move  our  hands  around  in  space, we  use  kinesthetic  feedback to  kind  of  locate  where  our  limbs  are. Kind  of  another  kind  of haptic  kind  of  type  is  cutaneous  or  tactile  feedback. And  this  deals  with  kind  of  the  skin  level  feedback. Things  dealing  with  temperature, texture  slip  on  the  skin,  vibration,  and  force. When  we  think  of  pouring  a  glass  of  water, some  of  the  kinesthetic  feedback  we  might have  are  the  weight  of  the  bottle  and  the  glass, of  the  orientation  of  our  wrist  in  space. While  the  cutaneous  or  tactile  feedback would  be  things  like  the  temperature  of  the  water, the  temperature  of  the  glass, kind  of  the  feel  that  you're  holding a  glass  and  not  a  paper  cup, and  also  kind  of  the  small  vibrations  as  you're  pouring, you'll  feel  like  that  trickle  and it  vibrates  the  tips  of  your  fingers. What  mediates  the  haptic  feedback in  our  skin  are  mechanorceptors, and  different  mechanorceptors  are able  to  sense  different  modalities  of  feedback. There's  things  like  pacinian  corpuscles  for  vibration, marco  cells  for  pressure, and  a  few  other  ones  that  sense  things  like  stroking, normal  force,  temperature,  et  cetera. These  mechano  receptors  are  also distributed  differently  throughout  the  body. Someone  who  works  on  a  lot  of  wearable  haptic  devices. This  is  also  a  really  interesting  area because  when  we're  looking  at  wearable  devices, they're  commonly  worn  where we  would  wear  a  watch  like  on  the  wrist. However,  in  these  areas, there's  actually  a  lower  mechano  receptor  density, making  things  to  design  around this  space  a  little  bit  more  difficult. When  we  think  of  haptic  devices  in  daily  life. Some  kinesthetic  devices  are  things like  desktop  devices  that  exist. Some  people  have  used  it  for surgical  simulators,  dental  simulators, where  you  move  a  pen  around  in three  D  space  and  there's  a  bunch  of  motors  that will  have  you  interact  in  three  D  space with  an  object  and  you  rigidly  touch  objects, and  you  can  feel  it  through  this  pen. Another  example  of  a  kinesthetic  device would  be  an  exoskeleton  or exost  where  it's  delivering forces  across  joints  in  the  body. On  the  other  hand,  cutaneous  devices tend  to  be  much  smaller  because  they  don't  need these  big  bulky  motors  to  act  across these  have  large  forces  to  act  across  joints. So  a  lot  of  these  wear cutaneous  devices  are  often  on  the  hand, where  we  have  those  higher  density  of  mecana  receptors. You'll  notice  that  there  are  also in  wrist  worn  or  arm  worn  devices  that  are  much  larger. A  cutaneous  device  that  a  lot  of you  encounter  in  daily  life  might  be your  smartphone  or  your  playstation  controller that  delivers  vibration. Many  fingertip  haptic  devices  can produce  salient  tactile  cues. But  as  mentioned  earlier, because  of  that  sparse  mechano  receptor  distribution, when  we're  building  displays on  the  arm  or  other  parts  of  the  body, it  can  create  a  few  complications. Because  of  this,  most  variable  tactile  displays on  the  arms  take  up  a  lot  of  real  estate. Um,  and  these  displays  have  been created  to  deliver  social  haptic  cues. So  things  like  haptic  emos, where  you  can  kind  of  deliver haptic  cues  that  elicit  feelings  of  happiness,  joy, or  sadness,  also  directional  cues, kind  of  for  path  following  path  navigation, consonant  articulation,  so  the delivery  of  language  through  touch. But  you'll  notice  a  lot  of  these  do  take up  that  large  amount  of  real  estate. Another  method  that  people  have investigated  is  actually  by  using  multi  modal  cues. Instead  of  just  having one  device  across  the  whole  arm  that delivers  different  patterns  of vibration  over  a  large  space. What  if  we  had  a  device  that  might  produce  vibration or  type  of  skin  shear  or  normal  indentation. Those  multi  modal  cues  would  in  theory  interact  with different  types  of  mechanic  receptors allowing  for  users  to  identify  different  cues. The  goal  of  this  first  project  was to  develop  a  arable  haptic  device  that can  simply  use  multiple  degrees of  freedom  for  feedback  on  the  farm. There  are  two  main  research  questions. The  first  is,  how  can  we  design  devices  to  add  more  cues? So  a  lot  of  shear  devices  were  one  directional. But  what  if  we  could  just  give different  paths  on  the  arm  and  say  like, nor  South  East  West  and all  the  intercardinal  at  45  degree  intervals? Also,  how  can  we  use  rigidity  and softness  in  designing  these  haptic  devices? Ideally,  when  we're  thinking  of  wearables, we  want  something  soft  and  compliant  to  be  worn. But  as  we'll  see  in  this  project, that  doesn't  always  play  out  that  well  and  that  you might  need  some  rigid  components in  addition  to  soft  materials. This  is  a  video  just showing  the  device  working  on  the  arm, and  we  created  this  device  that  uses  motors  and soft  actuators  to  deliver those  directional  shear  cues  at  the  wrist. Going  over  the  design  of  this  device. We  use  soft  actuators  that  are  constrained  with  fiber. These  are  fiber  reinforced  elastomeric  enclosures, and  they  create  all  the  linear  movement  on  the  device. There's  also  a  tactor  that  uses a  bubble  soft  actuator  that  goes  above  the  skin. All  of  these  soft  components create  all  the  linear  motion  needed. In  addition  to  this,  we  use a  rigid  motor  for  a  rotational  motion to  orient  the  direction  that  that  shear  will  occur  at. In  addition,  for  this  in  lab  test  device, we  integrated  force  sensors,  position  sensors, and  offboard  not  shown  here are  also  pressure  sensors  that measure  the  air  pressure  in  each of  the  pneumatic  components. That  was  done  to  characterize  how  the  device  is actually  working  on  the  skin  when  it's  delivering  forces. A  lot  of  haptic  devices  that  are  warrant  on  the  skin. They  don't  actively  measure  the  force  delivery, but  it's  most  of  them  work  on  an  open  loop  paradigm. I  will  say  this  device, we  did  the  tests  also  on  open  loop, but  we  wanted  to  measure  the  forces  to  see if  they're  consistent  in  how  they  were  being  delivered. I  won't  talk  about  the  data  too  much from  those  sensors  and  that  analysis, but  there  was  a  whole  subset  of  analysis  on  the  forces, kind  of  the  distance  that  each  kee  was  delivered, and  also  trying  to  understand  how  that matches  to  how  people  perceive  the  haptic  kes. Using  the  different  degrees  of  freedom, the  device  can  provide  multi  modal  heptic  cues. By  combining  the  normal  and  lateral  soft  actuators, we  can  create  shear. By  combining  normal  and  rotational, the  device  can  create  torsion. What's  happening  in  this  is that  bubble  soft  actuator  in  the  middle  can  bubble  out. It'll  touch  the  skin  and  create  a  twist  on  the  skin. You  can  also  add  vibration  to  any of  these  degrees  of  freedom  by  creating pulses  in  the  pneumatic  lines or  by  oscillating  the  motor. This  is  an  example  of  how  we  can have  more  traditional  vibration  queues that  might  be  more familiar  from  a  phone  and  this  could  be  used  for  alerts. To  test  the  device,  we  mainly  wanted  to analyze  how  the  shear  queues  on  the  device  worked. To  do  this,  we  did  a  user  study  that  we had  28  participants  doing  two  different  scenarios, a  four  direction  task, and  an  eight  direction  task. These  are  analogues  of some  previous  experiments  that  we  did  in lab  on  devices  that  didn't  work  differently. This  consisted  of  three  phases, a  training  phase,  a  practice  phase, and  finally  a  test  phase. For  the  four  direction  task, we  were  able  to  get  an  accuracy  of  85%. And  note  that  chance  for  this  case  is  25%. So  if  you  had  a  participant  guessing  randomly, you  would  expect  that  they  perform  at  about  25%. The  average  angular  error  was  19  degrees, and  shown  here  is  a  confusion  matrix. For  those  who  might  be  less  familiar, on  the  x  axis  is the  delivered  stimulus  on the  y  axis  is  what  is  perceived. In  an  ideal  case, you  would  have  one  hundreds  going  across  that diagonal  and  you  would  want  it  to  be  all  dark  blue. That  means  that  all  of  the  cues  that  were  delivered  by the  device  were  perceived  correctly  by  the  user. From  this  example,  we  can  see  that  a  lot  of the  confusion  mainly  comes in  queues  in  the  opposite  direction. If  we're  delivering  a  queue  that's up  like  in  the  first  column, it  means  that  about  12%  of the  time  they  perceive  it  as  going  down. You'll  notice  this  is  an  interesting  thing. Earlier,  I  talked  about  rigidity and  softness  in  these  devices. And  what's  interesting  is  that  why are  they  feeling  this  opposite  force? It's  because  the  device  has  to  be grounded  on  the  arm,  because  it's  a  wearable. When  you  think  of  the  forces  that  are  being  delivered, each  force  has  an  equal  and  opposite  reaction  force. So  when  you're  delivering  the  cue, the  device  actually  is  grounding  that  on  the  arm, so  you  get  that  confusion  in  the  force  delivery. In  the  eight  direction  task, as  expected,  you  would  have  a  lower  accuracy. This  is  43%,  which  is still  significantly  better  than  chant  at  12.5. In  this  case,  we  see  that  a  lot  of the  errors  in  identification  are at  angles  near  or  adjacent  to  the  delivered  stimuli. The  angular  error  is  49  degrees, so  that's  a  metric  that  shows  that  you're  within just  one  boundary  away  from  the  desired. So  in  summary,  we  created  this  wrist  worn haptic  device  that  uses soft  and  soft  actuators  and  a  motor, and  we  were  able  to  create a  shear  display  on  the  arm  that gives  these  linear  directional  cues. I  didn't  talk  about  some  of the  displacement  models  today, but  that's  in  the  paper. And  also,  we  can  look  at  and  interpret different  cues  from  the  device  an  effective  manner. After  doing  my  first  few  years  work  in  Haptics. I  also  I'll  talk now  about  this  massive  open  online  platform  for  Haptics. But  reminding  to  COVID  19, so  I  had  some  time  in  lab  building  these  Haptic  devices. But  as  I  went  home  during  COVID, talking  about  what  I  do  to  my  parents  and  they're  like, Yeah,  I  don't  know  what  that  is. What  are  you  even  doing? I  was  thinking  of  ways  that  we  could bring  it  into  the  community, teach  them  about  the  work,  and maybe  use  it  as  an  educational  tool. Another  thing  that  I  worked on  during  COVID  was  a  Scholars. Take  anything  from  this  panel. Please  take  this.  Come  home.  Come  home. There  are  jobs  for  you.  There  are  programs  from  you. We  need  you  all.  Come  back  to  Hawaii. COVID  is  going  to  be  our  long  term. Where  that  place  is  like your  presence  there  is  important  and  valuable. One  very  important  thing  which  I  didn't  do, which  I  highly  recommend  doing is  networking  and  communicating  with  people. Don't  be  afraid  to  ask  Fill. Don't  be  afraid  to  ask  your  resources. The  family  and  the  friends maybe  something  like  Hua  Scholars. It  is  completely  invaluable. That  was  what  kept  me  going through  different  things  when  you're  only. The  only  Coking  class. We  have  Stem  industry  here  in  Hawaii, and  that's  very  important  to know  that  there  is  a  job  for  you. Thank  you  so  much  for  Oua  scholars  and  for the  continued  success  of  this  organization, the  advancement  of  Stem  opportunities in  these  islands  and  for the  future  of  Hawaii  and  all  of  our  people. Cool.  That  was  like  our  little  commercial  snippet. But  essentially,  when  I  went  back  home  during  COVID, a  lot  of  people  were  struggling  to  keep up  in  school  with  remote  and  distance  learning. But  it's  not  always about  understanding  what  is  going  wrong, but  it's  about  how  can  we  use  COVID  as  an  advantage? Actually,  Hoa  scholars,  we  started  out as  an  online  stem  mentorship  resource. Actually,  we  realized  because  everyone's  getting  used  to Zoom  and  distance  learning  and  all  these  online  meetings, it  was  actually  a  good  way  to  loop Kanaka  who  are  not  in Hawaii  to  actually  support  people  in  Hawaii. It's  about  figuring  out  how  can  you use  things  that  are  unideal, but  actually  find  new  ways  to  benefit  the  community. Over  the  past  three  years,  we've  grown  it. We've  gotten  a  lot  of  sponsors  along  the  way. Now  we  hold  annual  symposiums. We  have  a  proposal  competition. This  year,  we're  giving  $5,000  awards  to  students. Over  100  students  in  Hawaii  apply every  for  the  past  few  years. It's  like  growing  out  this  new  ways of  thinking  and  new  ways  of getting  ideas  in  the  community. So  using  that  focus. I  also  learned  about  a  lot  of other  projects  in  the  community  that  were  crowd  source. There's  things  in  Hawaii  like the  King  Tides  project  where  they  have everyone  go  out  with  their  smartphones, take  pictures  of  shorelines, and  they  try  to  analyze  it  to  see how  climate  change  is  impacting  the  coast. There's  also  projects  that crowd  source  bird  watching  in  Hawaii  where there's  databases  of  birds  and they  track  the  frequency  that  birds  show  up. Um,  but  there's  all  of  these  crowd  source  tools, and  I  was  like,  Hey, what  if  we  could  implement  that  also in  the  work  that  we  do  in  Haptics? So  I  along  with  a  few  other  people  at  Stanford, we  created  what  we  term  as the  massive  open  online  platform  for  Haptics. The  premise  of  this  is  that we're  doing  these  crowd  source Haptics  experiments  that  people  can  do  anywhere. The  idea  is  a  participant  could  just  scan  a  QR  QR  code, go  to  a  website,  download  the  app, and  do  a  bunch  of  user  studies, give  us  information  as  researchers, but  also  could  help  them  learn tools  that  we  use  in  our  academic  realms. Yeah,  I  got  a  few  of  my  friends just  using  the  smartphone  and  walking  around  campus. There's  been  three  large  papers that  have  come  out  of  this  work. The  first  is  a  reaction  time  study, which  was  just  measuring  reaction  times  to different  audio  Haptic  and  visual  cues. Delivered  by  the  phone.  I  also  did  a  distraction  study, which  I'll  talk  more  about  today, and  then  also  clinical  tests where  we  can  use  phones  to  deliver vibrations  to  actually  monitor  peripheral  neuropathy. We're  actually  currently  working  with  doctors at  Stanford  and  getting a  larger  data  set  and  matching  these to  traditional  clinical  benchmarks. And  what's  really  interesting  about  all  of these  projects  is  that  in  Hawaii, where  there's  remote  communities or  people  on  different  islands, having  these  tools  and clinical  assessments  or  different  apps  for  learning, it  allows  people  in  these  more  rural  areas  to reconnect  with  some  of  the  things that  we're  working  on  at  the  universities. So  today,  I'll  talk about  one  of  these  in  a  hart  vignette, which  is  a  distraction  study. This  is  a  paper  in  transactions  on  Haptics. So  oftentimes  we  design  Haptic  devices  for  daily  life, and  a  lot  of  these  haptic  devices are  actually  only  tested  in  lab. So  there's  Haptic  devices  for outdoors  to  help  people  walk  who  are  blinds, or  for  steering  wheels, for  feedback,  kind  of  AR  VR  in  that  area, but  The  real  world  has  a  lot  of  distractions. It's  really  complex  and  challenging. We'll  see  if  the  video  loads. Yeah.  And  there's  kind  of  another  researcher  who talked  about  how  running  in the  wild  user  studies  is  inherently  challenging. So  in  Haptics,  we're always  building  these  custom  devices, and  even  when  we're  using  them  in  lab, it'll  break  every  few  participants and  we  have  to  redesign,  rebuild. So  what  if  we  could  use  the  phone  as a  platform  to  test  and study  these  concepts  in  the  greater  world? So  this  project  had  two  main  questions. The  first  is,  how  does  our  vibration  perception change  in  the  presence of  cognitive  and  physical  activity? And  the  second  is,  can  we  use smartphones  to  broadly  study  vibration  perception? Some  prior  work  in  terms  of  distractions and  how  we  perceive  vibrations  has  been  done. It's  been  found  that  when  you're  contracting  muscles, you  have  different  sensation  of  vibrations. Also  when  people  are  walking  around, they  have  different  ability to  sense  vibration  at  different  parts  of  the  body. Also,  in  some  pilot  studies, it  was  found  that  when  people are  trying  to  identify  haptic  signals, they  have  a  more  difficult  time  under  cognitive  load. To  study  this  concept, more  in  depth  in  a  foam  based  platform, we  created  a  smartphone  that  delivers a  vibration  response  task  and  a  shape  memory  task. The  vibration  response  task  consisted  of  delivering vibrations  at  nine  different  intensities with  ten  repetitions  each. The  user  simply  presses this  red  button  every  time  they  feel  a  vibration. So  you  can  imagine  you're  holding  the  phone, the  phone  is  just  vibrating  randomly, and  we're  just  like  press the  red  button  every  time  you  feel  a  vibration. Now,  the  shape  recall  task was  done  to  impart  cognitive  load, and  how  this  works  is  that  there  are four  different  shapes  shown  in  sequence, and  they're  shown  every  3  seconds. And  they're  shown  in  a  pattern  such that  we  ask  the  user  to  press this  blue  button  when  they  see the  same  shape  shown  consecutively. So  this  is  a  play  on  the  nba  task, which  is  commonly  used  to  impart  cognitive  load. So  for  example,  if  we  had  a  shape  sequence  of  square, circle  triangle  triangle  square, you  would  have  pressed  that  blue  button when  you  see  the  second  triangle. And  yeah,  the  shape  is just  shown  on  the  face  of  the  phone. In  addition  to  this,  we  worked  on  designing  a  phone  case that  can  be  used  for  these  user  studies for  the  in  lab  testing. The  idea  with  this  is  that  the  phone  case would  help  to  control  where  people  place  their  hands. We  took  vibration  accelerometer  measurements. The  phone  case  was  designed  so  that  no  one  could  squeeze the  phone  because  then  if they're  imparting  extraneous  forces, it  would  then  also  impact  that  vibration  perception. Um,  kind  of  alongside  this  study, as  I  mentioned  earlier,  we  are also  doing  that  clinical  study. So  we  wanted  to  create  other  stand  ways  of  holding phones  that  we  can  understand what  queues  are  being  delivered  to  the  user. In  this  experiment,  we  had  24  participants  with four  combinations  of  cognitive  and  physical  activity. Denoting  the  level  of  activity  is  lowercase  or  uppercase. In  this  top  left  corner  where a  participant  is  sitting  and simply  responding  to  the  vibrations, we  denoted  it  as  C  P.  We  also  had  a  condition  where the  participant  is  doing the  same  vibration  response  tasks, but  doing  it  on  a  treadmill  while  they're  walking. We  also  tested  if  they're  seated, but  then  they  have  that  N  A  shape  memory  task. This  is  the  case  where  they  have  low  physical  load, but  high  cognitive  load. Then  also  the  case  where  they're  doing the  shape  memory  task  while walking  and  responding  to  vibrations. In  this  case,  it's  high  cognitive  load and  high  physical  load. This  shows  example  data  from  one  participant, and  this  is  a  psychometric  curve. What  it  is  is  it's  just  a  plot  of  the  number of  the  proportion  of detected  vibrations  for  each  amplitude. On  the  x  axis, we  have  haptic  intensity, which  is  that  vibration  amplitude, and  on  the  y  axis, we  have  the  proportion  of  vibrations  that  were  detected. As  expected  in  the  bottom  left  hand  corner, you  would  have  very  few  vibrations detected  because  they're  harder  to  sense. Then  when  we  have  large  vibrations  delivered, we're  able  to  detect  most  of  those  vibrations. What  we  can  do  is  for  each  condition, we  can  plot  the  data  for  all  participants, for  each  participant,  and  then  fit  it  with  a  function. Then  using  where  it  crosses this  50%  detection  threshold indicated  by  the  black  dotted  line. We  can  see  how  the  different  conditions affect  their  perception. In  this  case,  we  see  that going  from  light  green  to  dark  green, this  is  a  case  where  they  have  increased  physical  demand. We  can  see  that  the  curve  shifts  to  the  right, indicating  that  they  need  larger  vibrations  to sense  kind  of  that  larger  vibrations  for  detection. That  previous  graph  was  just  for  one  participant. What  we  can  do  is  plot all  the  participants  on  this  graph  and  get  averages. In  the  yellows  and  reds  are  individual  subject  data, and  then  we  have  the  means  and standard  error  shown  in  the  purples  and  greens. Here  we  found  that  there  was  a  significant  effect  of both  cognitive  and  physical  activity on  vibration  perception. When  we  go  from  the  low  physical  activity to  the  high  physical  activity  state, we  see  that  there's  an  increase in  vibration  perception  threshold. And  similarly,  when  we  go  from low  cognitive  activity  to  high  cognitive  activity, we  also  see  an  increase  in  that  threshold. We  also  measured  how  response  time to  the  vibrations  also changes  between  the  two  conditions. Here,  we  actually  found  that the  main  difference  in  determining response  time  was  the  impact  of cognitive  load  that  the  user  had. Here  in  the  low  cognitive  conditions, they  have  this  shorter  response  time, but  when  they  have this  dual  task  assessment  with  the  shape  memory  task, they  actually  have  slower  response  times. We  also  found  that  participants  reported increased  demand  with  task  difficulty. This  shows  that the  mental  and  cognitive  demand  that  we  imparted  to the  users  actually  resulted  in an  increase  of  user  reported  mental  and  cognitive  demand. In  summary,  we  found  that cognitive  and  physical  activity can  impair  vibration  perception. Cognitive  activity  increases  response  time to  vibrations  and  that widely  accessible  smartphones  can be  used  to  study  vibration  perception. And  in  the  catch all  is  that  I'm  really  interested  in  working  with future  collaborators  on  how  we  can  bring other  apps  or  other  studies  into  this  platform, and  also  bringing  it  in  as  an  educational  tool, where  we  can  teach  people  about statistics  and  a  lot  of  these  other  tests  that  we  do. So  now  I'll  talk  about  some  work  in  soft  robotics. So  I'll  talk  about  this  fiber  reinforced  soft  actuator. This  was  a  robo  soft  paper all  the  way  back  during  COVID  2020. So  a  lot  of  times  we  use  these  rigid  robots  and  they  work very  well  in  predictable  controlled  situations such  as  in  manufacturing, but  they  also  pose  risk  to  humans. So  if  you  cold your  robot  wrong  or  the  human  goes  within  the  barrier, It  can  actually  cause  harm  to  humans. One  of  the  methods  that  nature  inherently  uses is  by  using  soft  bodied  mechanical  intelligence. Elephant  trunks,  they're  flexible. They  can  move  around,  do  a  variety  of  tasks. We  also  have  animals  that  can  fit  into really  small  spaces  by  using that  inherent  soft  body  compliance. And  soft  robots  are  one  avenue that  we  use  to  kind  of  mimic  this  area. So  we  have  robots  such as  McKibben  actuators  that  use  kind of  patterns  on  the  outside  so  that they  artificial  muscles  kind  of  constrain  them. Also  robots  that  actually  have  gait  and  walking. One  key  thing  is  that  there  are  these  class of  robots  called  fiber  reinforced  elastomeric  enclosures. This  builds  off  those  Mcibbin  muscles  where  you  use these  constraint  patterns  to dictate  how  an  elastomer  deform. What  this  is,  it's  just  silicone, and  then  you  wrap  strings  around  it, and  then  you  can  cause  it  to  extend, twist,  expand,  or  bend. However,  similar  to  rigid  robots, a  lot  of  them  are  one  robot  for  one  application. You  have  one  robot  that  can  flip  pancakes  really  well, but  it  doesn't  do  laundry. What  if  we  could  have  soft  robots  that can  change  their  output  and  do  different  things? The  idea  was,  can  we  create a  reconfigurable  soft  actuator  that  can produce  all  of these  different  multiple  output  configurations. How  can  we  actively change  the  configuration  of  these  soft  robots? Then  how  can  we  actively  change these  fiber  patterns  to  dictate  those  outputs? So  to  do  this,  I  designed this  active  fiber  reinforced  elastomeric  enclosure, which  uses  active  fibers  that are  constrained  around  the  soft  actuator. How  this  works  is  that  there's  two  motors  that independently  control  two  sets  of  fibers. So  there's  two  motors  shown  on  the  bottom. And  then  we  also  have  a  spool  at  the  top because  as  you're  changing  fiber length  around  the  actuator, that  string  length  will  also  change. Then  finally,  in  the  middle  is  that  soft  actuator, which  has  a  pneumatic  input  for  that  actuation. Here  we  can  see  that  when we  use  these  different  fiber  patterns, the  actuator  can  actually  produce  different  outputs. Now,  the  idea  is  like you  could  create  this  actuator  that  has multiple  output  configurations  that can  be  used  in  a  variety  of  scenarios. To  characterize  this,  we just  pressurized  the  actuator  and  tested the  different  fiber  angle  combinations and  measured  the  twist  and  displacement. We  also  had  two  different  conditions, one  where  we  spooled the  slack  in  the  fibers  and  one  where  we didn't  with  the  premise of  how  can  we  simplify  the  design? This  is  data  shown  at  the  highest  pressure, and  we  found  that  the  actuator  can create  varied  angles  of  twist  from plus  to  -60  degrees with  two  to  four  millimeter  displacement. As  expected,  there's  a  predominantly  symmetric  response because  the  fibers  are  symmetrically  patterned. Further,  we  can  I'll  point  out  a  few  things. So  we  can  see  that  in  the  bottom  left  and  top, right  corners  of  each  of  the  graphs, we  have  the  maximum  twist, which  also  coincides  with  the  maximum  displacement, which  is  expected  because  those  fiber  angles  are  aligned. So  in  the  case earlier  with  those  examples  with  the  fixed  fibers, there  are  fibers  that  do  predominantly twist  and  they're  kind  of  candy  cane  strike. In  those  cases,  you  would  expect  maximum  twist, and  then  when  they're  cross  pattern, like  a  double  helix, then  you  would  expect  minimal  twist, and  we  similarly  saw  that  in  these  plots. We  also  found  that  there  is  a  point  at  which the  fibers  activate  and control  the  response  of  the  actuator. We  took  some  dynamic  response  graphs, and  we  found  that  at  different  pressures, you'll  get  different  steady  states  based  on  how the  soft  actuator  is interacting  with  the  fiber  configuration. To  summarize,  we  found  that these  active  fiber  constraints  can increase  the  range  of  motion  and create  other  patterns  that  one  to one  mappings  of  robots  to  applications  traditionally  do. And  currently,  we're  working  on coupling  these  actuators,  reducing  the  size. Our  current  one  actually  only  uses  two  motors  now, so  we  found  ways  to  reduce  the  complexity  of  the  design. Yeah.  So  After  designing these  soft  robotic  devices  inside  of  lab. It  also  made  me  interested  in  kind  of  the  area  of collaborative  robotics  that  I  know a  lot  of  people  here  might  be  working  on. And  we  often  see  that  robots are  not  well  integrated  into  our  communities. So  to  the  extent  that  a  lot  of  people  will see  robots  is  through  their  vacuum  cleaner. But  we  also  have  robots that  can  do  a  lot  of  tasks  currently, like  manipulation,  grasping  objects. This  could  be  really  critical  in home  environments  or  in  senior  care  facilities. So  I  think  there's  this  huge  potential  for  growth. There's  a  lot  of  reasons  why  we  don't use  these  traditional  rigid  robots  in  daily  life. Some  of  it  stemming  from  the  technology itself  and  the  dangers  that  it  poses, but  there's  also  these  psychological  constraints. Fear  of  obsolescence,  if  you're  trying to  design  these  robots  along  with  workers. I  recently  heard  of  a  story  where  there is  these  robots  that  are  used  in hospitals  and  nurses  try  to sabotage  the  unit  because  they  don't  want  to  be  replaced. I've  also  seen  in  some  airports they  have  robots  that  pick  up trash  and  people  will  intentionally  just step  in  front  of  it  just  to  piss  the  robot  off. But  there's  a  lot  of  these  kind  of  things  that  people do  when  they  encounter  new  technologies. And  soft  robots  actually open  up  a  really  different  avenue  where soft  robots  kind  of  because  of that  biological  inspiration  and  compliance, they  might  actually  be  ideal for  collaborating  with  humans. This  soft  robot,  it's  called  the  Vn  robot. It  came  out  of  the  PhD  lab  I  worked  in  at  Stanford, but  essentially,  it's  this  everting  tube. Where  you  can  use  air  pressure  to evert  this  tube  and  grow, so  I  can  do  a  variety  of  things  and grow  through  really  small  spaces. The  best  part  is  it's  pneumatic  and  it's  made  out  of fabric  or  this  low  cost  plastic. Working  along  that. We  use  similar  concepts  to  build these  fabric  based  robot  manipulators that  can  actually  be  used  for  collaboration. Currently,  we're  working  on  it  in  a  few  areas. But  essentially,  we  can  use  these  concepts  in soft  robotics  to  design  new  types  of robots  that  interact  with  people  in  daily  life. In  agriculture,  we're  working  on  new  soft  grippers. We  have  a  soft  robot  system  that can  go  and  grow  out  towards  apples  and  pick  apples. Some  of  my  current  work  at  Washington. Then  also  understanding  these  psychological  aspects of  human  robot  collaboration. Compared  to  this  traditional  rigid  linkage  robot, what  if  we  replaced  it  with a  soft  bodied  robot that  would  pose  less  risk  to  the  human? How  can  we  control  those  robots  to  have better  aspects  for  collaboration by  changing  speed,  proximity? Does  it  work  in  a  cooperative  or  collaborative  manner? So  those  are  some  of  the  questions  that I'm  currently  working  on  and  tackling, and  it  shows  kind  of  how bringing  out  the  primary  research  in the  robot  itself  makes  you  think of  how  the  robot  can  be applied  to  these  different  communities. As  an  outline,  I  covered a  few  different  topics  in  Haptics  and  soft  robotics, and  then  also  how  I  use  some  of  my  background  to bridge  those  areas  into  more  community  facing  projects. Altogether,  this  forms  like some  of  the  future  collaborations that  I'm  interested  in  in the  Realm  soft  robotics  Haptics  and that  community  based  education  approach that  we  can  use  in  research. I'd  like  to  acknowledge funding  resources  for  different  projects, Um,  I've  worked  with  a  huge  I  think  it's a  very  large  number  of  people  throughout my  PhD  and  at  the  beginning  of  my  post  stock. But  shout  out  to  a  lot  of the  really  strong  undergraduates who  made  a  lot  of  the  work  possible. And  I  really  enjoyed  mentoring a  lot  of  these  students  and  working  alongside  with  them, and  also  my  PhD  advisor  Allison  and  Posto  advisor  Ming. And  I  think  yeah, that  is  my  last  slide. So  yeah,  I'm  happy  to  take  any  questions. And  thank  you  for  having  me. Questions.  I  have  a  mice  if  people want  to  use  it  or  to  yell  out  a  lot  of  people. I  do  have  a  few  that  I  can  kick  us  off  if  you  want. Yeah.  Kyle,  one  thing  I  noticed, and  I  really  appreciate  the  talk. And  how  you  we  some  of  your  own  journey  and  story  india. There's  a  I  say different  types  of  students  in  the  room  too, and  your  presion  move  back  in  one  sense. But  I  think  a  lot  of  students  things  I  can  ask  are how  do  you  do  the  scientific, technical,  academic  work  and  then also  get  back  or  work  with  intrigue  to  your  community. I  have  an  example.  But  how  did you  think  of  that  when  you  were  first  starting  on? What  was  best  question. So  I  When  I  was  in  undergrad, I  went  on  this  retreat, and  there  was  like  a  business  school  professor who  gave  a  p.  Kind of  so  we've  all  come  from  like cultural  backgrounds  with  native  wai  values. But  also,  like,  I'm  part  japanese. And  I  didn't  know  of  it, but  there's  this  concept  called  P which  is  kind  of  the  balance  of  what. It's  like,  what  you  like  to  do, what  the  community  needs,  what  you're  good  at. I'm  on  what  you  want  to  do. But  there  is  kind  of  I  don't  believe. As  I  talk  to  other  people, I  think  about  money  isn't  a  factor  in what  I  do  or  what  I  should  do, because  I  think  as  long  as  we  focus  on  the  three  areas, what  we're  good  at,  what  the  community needs.  I  forgot  the  third. But  as  long  as  you  focus  on  those  three  areas. I  could  start  the  list  again,  but  I  just  said  it. As  long  as  we  focus  on  those  three  areas, so  money  will  always  come  after. So  as  a  student. I  was  always  thinking  of,  like,  Oh,  like, I  really  love  kind  of  some  maths. Is  too  much  math, but  some  math  things. So  my  engineering  and  connecting  back  like, Oh,  how  can  I  use  this  for  the  love  of  community. While  I  also  accelerating the  body  of  research  that  I'm  working  in. So  I  think  about  as  a  student, no  matter  where  you  are,  high  school  undergrad  student. There's  always  a  project  that you  might  be  actively  working  on  now, but  it's  also  how  can  I use  those  skills  to  work  on  something  that the  community  or  I might  be  interested  in  as  a  side  project. So  actually  pal  of  those  px. The  clinical  test,  it  was  all  It  all started  off  as  a  psych  project  and  my  was  like, you  don't  file  like  it's  you  can  b. You  should  try  to  write  the  graph  and  see  what  happens. Eventually  between  this  large  section of  the  lab  and  what  I. So  I  just  pursuing that  passion  while  using  existing  skills, I  think  it  is  a  great  thing  a  practice. Any  other  questions? Thanks  for  presentation. I  thought  research  topic  is  related  to  hepta  feedback. So  have  you  considered any  passive  hepta  devices for  similar. Do  you  have  any  experience  on  that? So  are  great  questions. So  one  of  the  people  who  were  Katy. She  was  a  full  the  PhD. But  she  actually  helps  re  a  lot  with  these  projects. S  actually  does  a  passive  learning. So  I  haven't  touched  into  that  specifically, but  I  am  interested  in  kind  of these  passive  habit  solutions, where  it's  not  too  much  in  your  face  habit. So  there's  a  project  that  I'm  working  on  now  for, like,  bias  of  robots  using  copic  devices. But  focus  on  do people  like  detect  how do  they  respond  to  the  pits  itself? How  do  you  use  the  bits  to  try to  train  them  to  do  cyt  motions. And  I  think  that's  kind  of  a  whole  another  area  that  is a  little  near  to  me  that  I'm really  interested  in,  but  yeah. That's  the  best  idea. A  lot  of  the  systems  that  we  build are  systems  that  are  unocive, and  kind  of  like  naturally  go  in  what  we  are So  yeah.  Do  you see  soft  body  robots  changing your  perception  in  regards  to  the  fear  of  obsolescence? Because  I  know  you  talked  about  the  fear  of  harm, but  I  don't  know  if, like  the  design  looks far  more  approachable  than  something  rigid. Do  you  see  that  like  combating  that  fear  as  well? For  human  obsolescence. Yes.  So  for  fear  of  human  obsolescence. Found  something  that  I  haven't  tackled  yet. I  will  say  that  recently, I  did  a  kind  of  a  trip  to  Hawaii  to  see  how we  can  design  your  robots  for  things  in  agriculture. And  a  lot  of  people  are  like,  punch. It's  really  important  to  have  the  kids  come  out and  actually  work  understaff  all  the  loves. And  I  100,000%  degree. I  think  keys  in  using  robots. It's  how  can  we  have  them  work  alongside  like  human. At  Allegion  explicitly  replaced  like  human  immediately. So  in  the  realm  of  kind  of  the  ark  now  in  Washington. There  is  a  huge  labor  shortage where  there  aren't  enough  people  to  go  and  pay. I  kind  of  arts  is  diminishing  over  time, and  especially  as  climate  change  is  coming, there's  a  lot  of  a  lot  more harsh  conditions  for  the  workers. So  But  to  see. We  have  been  working  for  30  years  on  te  robots. We  haven't  found  a  way  to  fully  automate  everything. So  the  curry  that  we're  working  on  is  actually trying  to  see  how  we  can  have these  robots  work  alongside  humans. So  maybe  there's  fruit  that  are  harder to  reach  or  more  difficult  to  grow. And  how  do  we  the  robot  work  alongside  the  humans. I  think  things  like  healthcare,  also. How  can  we  have  them  alongside  a  healthcare  worker? So  like  the  robotic  carts  that  go  around  furnaces. It's  like  I'm  sure  heart  cannot  really  replace  your  job. Like,  we  definitely  need  you, but  there's  still  kind  of,  like, things  This  feeling  against  certain  things  in  robots. I  think  is  to  have kind  of  a  design  process integrate  with  community  and  people  you're  working  with. And  have  them  build their  robots  or  have  the  build  it  with  you. And  also  looking  at areas  where  we  can  have  them  co  exist. I  think  ultimately  I  don't think  I  think  in  a  lot  of  cass, we  won't  fail  to  explicitly  replace, but  I  think  in  most  pss  how  can we  work  alongside  it  and  ease  our  working? Can  robots  together? I  guess  this  is  a  question  about  the  future  of  Haptics. I  feel  like  you  went  from  a  word  digital  appliance  or digital  system  output  where people  were  spurring  hept  options, but  nobody  really  had  had, had  any  of  them  to this  exposure  of  all  of  us  having  the  vibrating. Digital  appliance.  And  it  a  very common  output  modality  for  digital  appliances. And  I  just  often  have. Is  it  the  right  atic  output  for  all  of  us? And  is  there  a  future  where  there might  be  something  or  something that  would  be  quite  different  maybe richer  than  the  vibration, which  really  only  has, you  know,  you  can  pulse, you  can  get  two  vibration  versus  one  or  three, but  it  feels  fairly  impoverished. So  I  guess  I'm  wondering  where  the  future  of  tics. Will  be  my  phone. Simple  question. I  was  a  great  question. A  lot  of  you  guys  might  have  seen like  metals  that  came  out. I  think  Mark  Suckerberg  has  a  really  cool so  like  new  mil  and  you  can  grab  things  in  spits. But  along  these  devices  that  are  kind of  like  technology  like highlights  show  where  the  field  is  I  think  it's  like still  and  I  and  a  very  early  research  stage. A  lot  of  those  devices  right  now  are  super  expensive. So  there's  kind  of  ideas  of how  we  can  have  these  soap  devices. That  can  be  right  next to  the  computer  that  we  can  use  all  the  time. But  I  think  that  in  the  future, we  will  move  kind  of beyond  simple  vibrations  from  phones. I  think  that  it  would  take a  lot  more  work  because  of  the  energy  use. So  kind  vibration  waters,  they're  really  small, but  to  deliver  like  sheer  the  Marvel  haptic  devices are  arable  devices  that might  and  other  parts  besides  the  Hinds, that's  kind  of  like  the  force  output. I'll  use  more  energy. It'll  also  here. So  I  think  there's  work  in  kind  of  the  materials, and  energy  side,  but  also  in  the  optic  perception  side. But  yes,  I  think  there  is  there  are pathways  to  move  beyond  traditional  phone  vibration. So  other  areas  that  I  think  are  very appealing  are  paper  based  optic  devices. So  because  papers  can  be kind  of  machine  kind  of  rod  role  manufacturing, really  easy  to  mass  produce. I  think  that  might  be  one  of the  intermediate  transition  stuff, you  might  see  it  I  won't give  the  years  because  I  might  get ca  I'm  really  interested  in so  fabrics  might  be  the  next  step  for  normal  indentation, very  simple,  so  low  cost. I'm  going  beyond. That's  a  very  provoco  question. Imus  to  see  what  you  think  of  the  future. Awesome.  Well,  over  almost  the  time, to  make  sure  I  remind  you  all again  to  take  your  trash  and  put  it  in  the trash  g.  Let's  take  your  time. T  time  and  he'll  be  up  here  for  a  few  minutes, if  you  want  to  come  by  and  say,  Hey, but  otherwise,  tanks  to  come.