Skip to content
BinEnQ — for a sustainable and responsible tomorrow

Technology

Physical AI, built for the real world.

BinEnQ connects computer vision, edge intelligence, physical hardware and cloud services into one integrated learning system.

System architecture

From student to dashboard.

  1. Student
  2. NFC identification
  3. Waste presentation area
  4. Camera
  5. Edge intelligence
  6. AI detection / classification
  7. Student category decision
  8. Interaction engine
  1. Hardware controller
  2. Actuator
  3. Correct bin
  4. Local event record
  5. Secure synchronisation
  6. Cloud platform
  7. Gamification + analytics + dashboards

Six technical layers

A layered platform.

1Physical Interaction
Identification, waste presentation, category-selection controls and physical compartments.
2Hardware Control
Microcontroller-based control of actuators and lids, with sensor confirmation of the actual physical state.
3Edge Intelligence & Device Software
Camera input, AI inference, interaction logic, local event recording and offline resilience.
4Cloud & Application Services
Secure synchronisation, device management, configuration and application services.
5Data, Analytics & Gamification
Structured learning events, aggregation, analytics and gamification rules.
6User & Stakeholder Access
Role-based dashboards for students, teachers, schools, authorised public-sector stakeholders and product operations.

Edge-first architecture

Intelligence where the interaction happens.

The core interaction is designed to operate locally wherever practical, rather than requiring a cloud roundtrip for every interaction.

Local functions may include

  • Student identification
  • Camera input
  • Waste detection
  • AI inference
  • Category selection
  • Evaluation
  • Feedback
  • Lid control
  • Local event recording

Cloud services support

  • Synchronisation
  • Dashboards
  • Analytics
  • Device management
  • Gamification
  • Cross-device functionality

Edge-first path

  1. Student
  2. Device
  3. Local AI
  4. Hardware

Responsibility split

AI understands. The control system acts.

AI does not directly control actuators.

AI Layer

“What is the item?” → classification result

Device Software

“What should happen?” → interaction decision

Hardware Controller

“Perform the physical action.”

Sensors / Feedback

“Confirm what actually happened.”

Hardware

The BinEnQ system, component by component.

BinEnQ concept render showing the camera above the display, NFC card reader, waste presentation tray and five physical waste-category compartments labelled plastic, paper, wet waste, metal and general
Concept render. Students choose the waste category with physical controls; the screen is used for instructions, AI status, feedback and gamification.
display / feedbackwaste detection areaphysical category buttonscompartmentcompartmentcameraNFC
  • Camera
  • NFC reader
  • Display
  • Waste detection area
  • Physical category buttons
  • Edge computing unit
  • Microcontroller
  • Actuators
  • Lid feedback sensors
  • Compartments
  • Sensors
  • Connectivity
  • Local storage
  • Cloud synchronisation

Product operations

Devices, AI monitoring and platform management.

Platform Operations

BinEnQ product administrator view

Illustrative View
Registered schools

34

Devices

118

112 online / 6 offline

Interaction volume

9,120 / day

System alerts

3 open

AI monitoring

Uncertain classifications5%
Retry / re-present events8%
Completed interactions87%

Model version and device software version are visible per device.

Device events

  • Lid actuator confirmation missing — device 042 (investigating)
  • Connectivity interruption — device 017 (synchronised later)
  • Software version rollout — 96% of fleet
  • Configuration update — gamification rules, region B
User & role management

RBAC

Authorised scopes

Gamification configuration

Per programme

Reports

Operational

Operational views avoid unnecessary student personal information.

Technical stack

Proposed / current baseline architecture.

Technology choices are a working baseline for the MVP and are not presented as permanently final.

The next generation shouldn't just know about sustainability.
They should practise it.

BinEnQ turns an everyday action into a learning opportunity — connecting Physical AI, gamification and sustainability intelligence to help build better waste-segregation habits.

Become a Pilot Partner Experience BinEnQ