Integration and Validation

Integ­ra­tion is a very import­ant step, neces­sary to cre­ate DECENTER plat­form. 

This task is com­posed in two part:

  1. Integ­ra­tion: this activ­ity will integ­rate the advanced solu­tions invest­ig­ated dur­ing the pro­ject year between them and will cre­ate the DECENTER plat­form that will be tested and val­id­ate in the next step.
  2. Val­id­a­tion: this activ­ity is fun­da­ment­al to check the work done and under­stand how to improve it.

In DECENTER pro­ject there are 10 sci­entif­ic and tech­nic­al object­ives that will be reached:

  1. Define resource mod­els and asso­ci­ated SLAs for in-bor­der and cross bor­der ser­vices
  2. Devel­op a robust multi-tier fog com­put­ing plat­form to man­age cloud-to-edge resources
  3. Define and imple­ment resource orches­tra­tion strategies to sat­is­fy applic­a­tion require­ments
  4. Imple­ment block­chain-based smart con­tracts and prob­ing mech­an­isms to val­id­ate their ful­fill­ment
  5. Pri­vacy-pre­serving mech­an­ism for rep­res­ent­ing users’ con­text based on IoT data inter­pret­a­tion
  6. Facil­it­ate inter­op­er­able IoT device man­age­ment in the DECENTER infra­struc­ture
  7. Devel­op hier­arch­ic­al meth­ods to map AI algorithms on to the cloud-to-things con­tinuum
  8. Devel­op applic­a­tion data man­age­ment on dis­trib­uted infra­struc­ture for cross-bor­der AI
  9. Integ­rate and val­id­ate DECENTER innov­a­tions, and demon­strate them in real world pilots
  10. Pro­mote the stand­ard­isa­tion and indus­tri­al applic­a­tion of the DECENTER res­ults

Most of this object­ives will be demon­strated with dif­fer­ent use case imple­ment­a­tions.

Dif­fer­ent part­ners will lead and con­trib­ute to each use case demon­stra­tion.

USE CASE 1 – SMART CITY CROSSING SAFETY

The object­ive of this Use Case is to increase the ped­es­tri­an cross­ing safety by lever­aging the IoT and Edge infra­struc­tures in a smart city. This approach focuses on equip­ping a num­ber of ped­es­tri­an cross­ings with IoT devices exist­ing on the mar­ket, that enable mon­it­or­ing of ped­es­tri­ans intend­ing to cross the road. The object­ive of this Use Case is to spot any poten­tial dangers that might be put­ting the ped­es­tri­ans’ safety at risk and provide means for noti­fic­a­tion alerts.

DECENTER FEATURES USED

In this use case will be used the fol­low fea­tures from DECENTER’s archi­tec­ture:

  • Resource orches­tra­tion (Ver­tic­al)
  • Pri­vacy pre­serving AI
  • AI Mod­el repos­it­ory
  • Hier­arch­ic­al /distributed AI
  • Digit­al twin
  • Multi-tier fog com­put­ing plat­form
Ped­es­tri­an cross­ing in Vela with Decen­ter set up
Ped­es­tri­an sig­nage
Cam­era and micro­phone

USE CASE 2 – ROBOTIC LOGISTIC

The goal of this UC is to cre­ate a new lay­er of inform­a­tion that crosses from the side of the robots to the side of  Fleet Man­age­ment Sys­tem (a cent­ral­ized sys­tem man­aging and dis­patch­ing all the move­ments and actions of the robots), to add new inform­a­tion to the sys­tem, to improve the man­age­ment of the fleet and also to provide new inform­a­tion related with the status of the ware­house.

DECENTER FEATURES USED

In this use case will be used the fol­low fea­tures from DECENTER’s archi­tec­ture:

  • Resource orches­tra­tion (hori­zont­al)
  • Mod­el repos­it­ory
  • Secur­ity and com­pu­ta­tion off­load­ing (hori­zont­al and ver­tic­al)
  • Digit­al Twin

At the moment the devel­op­ment of robot­ic logist­ic pilot is focus­ing into the sim­u­lated envir­on­ment demon­stra­tion but in year 3 of the pro­ject we will move into real-world test­ing in Y3.

Web cli­ent (Gzweb) for the Gazebo-based robot sim­u­la­tion envir­on­ment

USE CASE 3 – SMART AND SAFE CONSTRUCTION APP

In order to achieve inform­a­tion integ­ra­tion for the above scen­ari­os, it would be neces­sary to imple­ment a meth­od­o­logy and a sys­tem that turns a spe­cif­ic con­struc­tion site into a smart and safe con­struc­tion site, by includ­ing a spe­cif­ic num­ber of cam­er­as and sensors on the ground, poten­tially, also on the act­ors (such as con­struc­tion work­ers, vis­it­ors) and objects (such as build­ing equip­ment, waste, mater­i­als etc.). Arti­fi­cial Intel­li­gence meth­ods will be applied on video streams in order to detect spe­cif­ic objects and use the inform­a­tion to issue noti­fic­a­tions.

DECENTER FEATURES USED

In this use case will be used the fol­low fea­tures from DECENTER’s archi­tec­ture:

  • Resource orches­tra­tion (Ver­tic­al)
  • Simple AI (not hier­arch­ic­al)
  • Digit­al twin
  • Multi-tier fog com­put­ing plat­form
  • Gen­er­ic AI con­tain­er­isa­tion mod­ule
  • Mon­it­or­ing ser­vice with multi-level met­rics
  • Markov decision pro­cess con­trol­ler imple­ment­a­tion for Kuber­netes
  • While pri­vacy pre­serving AI
  • Digit­al twin
Pla­cing objects and per­sons on a 2D floor plan or 3D mod­el
image detec­tion in a con­struc­tion site

USE CASE 4 – AMBIENCE INTELLIGENCE FOR SAFETY AT HOME AND AROUND

Vari­ous dan­ger­ous and uncom­fort­able situ­ations arise indoors par­tic­u­larly with depend­ent people such as eld­ers, dis­abled people and chil­dren. It is not easy to pre­dict and cope with these situ­ations unless a man­ager (i.e. par­ents, guide) con­tinu­ously mon­it­or them. Many IoT-based ser­vices are used for real-time safety mon­it­or­ing; how­ever, the focus of cur­rent ser­vices is mostly restric­ted to remote mon­it­or­ing of the cur­rent situ­ation with devices such as mobile phones.

This use case will show­case key fea­tures of DECENTER based on an AI applic­a­tion for indoor safety. This use case applies the out­come of DECENTER to the AI applic­a­tion to enhance detec­tion and alert of a dan­ger­ous situ­ation. 

 

DECENTER FEATURES USED

In this use case will be used the fol­low fea­tures from DECENTER’s archi­tec­ture:

  • Dis­trib­uted AI
  • AI mod­el repos­it­ory
  • Mul­tiple AI mod­els use
  • AI Ser­vice com­pos­i­tion with DECENTER facil­it­ies

UC4 pilot con­fig­ur­a­tion