PaperlayerView exact revision on arXiv

Source-linked AI summary

SmartBike: A Low-Power Edge IoT System for Cycling Safety and Real-Time Awareness

Sogand Hashemi, Gonçalo Sarabanda Neves Loureiro, Catarina Pereira De Freitas Lima, Lyssa Ramaut, Liesbet Van der Perre

arXiv:2610.11824v1eess.SP

TL;DR

Cyclists in dense traffic face limited situational awareness and a lack of active safety assistance, while practical bicycle platforms must operate for long periods on limited battery power. SmartBike combines event-driven sensing with lightweight edge-to-gateway monitoring and compact BLE broadcasts. Its estimated lifetime is approximately 21 months under typical commuting conditions, with sensing accounting for more than 90% of energy use.

  • Problem

    Cyclists remain vulnerable in dense traffic because conventional bicycles provide limited situational awareness and lack active safety assistance, while battery-powered platforms must balance sensing and communication against long operation.

    Source: §I · p. 1

  • Method

    SmartBike combines bicycle-mounted sensing and lightweight edge processing with piezo-triggered wake-up, application-specific sensing rates, and compact BLE broadcasts to a gateway.

    Source: §I · p. 1, A. System Architecture and Wireless… · p. 2, C. Low Power Design Considerations · p. 2

  • Results

    Approximately 21 months of estimated lifetime at 40 minutes of daily use with a 12,600 mWh battery; sensing accounts for more than 90% of consumption, while BLE transmission contributes a minor fraction.

    Source: §IV · p. 4

  • Takeaways & Limitations

    The prototype demonstrates cyclist-awareness functionality on a compact battery-powered platform, with sensing-side optimization more consequential for extending lifetime than further radio-overhead reduction.

    Source: §IV · pp. 4–5

    The authors identify external powering and long-range BLE support as remaining limitations.

    Source: §V · p. 5

Abstract

from arXiv · show

Urban cycling is a sustainable mode of transportation, but cyclists remain highly vulnerable in dense traffic because of limited situational awareness and the lack of active safety assistance on conventional bicycles. This paper presents SmartBike, a low power Internet-of-Things (IoT) platform for cyclist safety, combining blind spot monitoring, rider usage detection, localization, and wireless edge-to-gateway communication. The proposed system consists of a bike-mounted IoT node and a gateway assisted monitoring interface. The node integrates three time-of-flight sensors for blind spot detection, an IMU module for motion sensing and on-board speed estimation, a GNSS module for localization, and a saddle mounted piezo-electric transducer for event-driven wake up. To satisfy the energy constraints of the IoT node, the design employs piezo-triggered activation, application specific sensing rates, and compact connectionless Bluetooth Low Energy (BLE) advertisements for low overhead transmission. The firmware is implemented in Zephyr RTOS using a lightweight state machine architecture, while the gateway performs passive BLE scanning, payload decoding, and real-time dashboard visualization. The BLE payload is reduced to 11 bytes, and power analysis shows that sensing accounts for more than 90% of the total energy budget, while wireless communication represents less overall consumption. With a 12600 mWh battery, the estimated lifetime reaches approximately 21 months under typical commuting conditions (Average 40 min per day).

I. INTRODUCTION

Dense traffic leaves cyclists vulnerable because of limited visibility and a lack of active safety assistance, while battery-powered bicycle platforms must balance sensing and communication against compactness and long operation. SmartBike addresses these constraints with event-driven sensing and gateway-assisted monitoring.

I. INTRODUCTION

  • Cyclists face elevated accident risk in dense traffic because of blind-angle exposure, limited visibility, and a lack of active safety assistance on conventional bicycles,.The introduction also notes growing interest in intelligent sensing and low-power edge architectures for vulnerable road users,, and practical two-wheeler monitoring.Source: §I · p. 1
  • A practical bicycle platform must remain compact, lightweight, robust, easy to deploy, and long-lived on a limited battery budget.Always-on sensing and connection-oriented links may consume too much power, motivating joint optimization of sensing, processing, and communication.Source: §I · p. 1
  • SmartBike combines blind-spot detection, rider-use sensing, localization, and gateway-assisted monitoring using event-driven activation and application-specific sensing rates.The bike node performs lightweight processing and compact BLE transmission, while the gateway passively receives, decodes, and visualizes data in real time.Source: §I · p. 1

A. System Architecture and Wireless Data Path

SmartBike divides sensing and lightweight processing between a bicycle-mounted node and a gateway that receives and visualizes data. The node broadcasts compact, connectionless BLE advertisements rather than relying on persistent connections.

A. System Architecture and Wireless Data Path

  • The node acquires sensor data and performs lightweight local processing, while the gateway listens for BLE advertisements, decodes payloads, reconstructs physical quantities, and visualizes them.The division keeps packet reception, aggregation, and visualization on a less constrained platform.Source: A. System Architecture and Wireless… · pp. 1–2, Figure 1 · p. 2
  • The gateway-assisted partition preserves safety-monitoring responsiveness while avoiding unnecessary computational and communication burden on the bicycle node.The architecture treats the wireless link as a core part of the system design.Source: A. System Architecture and Wireless… · p. 2
  • SmartBike broadcasts compact, non-connectable BLE advertisements instead of maintaining persistent connections.This fire-and-forget model reduces connection overhead and radio-active time, and lets one or more gateways receive broadcasts without a return link.Source: A. System Architecture and Wireless… · p. 2

B. Hardware and Sensing Strategy

The sensing hardware combines a low-power BLE-capable controller with ToF, inertial, GNSS, and saddle-mounted piezoelectric sensing. The choices balance monitoring functions against power, installation complexity, and deployment convenience.

B. Hardware and Sensing Strategy

  • The XIAO nRF54L15 serves as the central controller, selected for BLE support and very low deep-sleep current.The hardware design balances sensing utility, power consumption, integration complexity, and cost.Source: B. Hardware and Sensing Strategy · p. 2
  • Three VL53L1X ToF sensors monitor the cyclist’s left, rear, and right regions for nearby vehicles or obstacles.They provide compact short-range proximity sensing for blind-spot awareness using optical-pulse round-trip time.Source: B. Hardware and Sensing Strategy · p. 2
  • An LSM6DSOX IMU supports motion sensing and on-board speed estimation without wheel-mounted instrumentation.It combines a three-axis accelerometer and gyroscope; relying only on GNSS-derived speed would use a relatively energy-expensive modality.Source: B. Hardware and Sensing Strategy · p. 2
  • An L76K GNSS module supplies position for ride tracking and real-time monitoring, but is treated as part of a broader power-aware sensing strategy.The passage characterizes localization as useful for environmental awareness but relatively costly in energy.Source: B. Hardware and Sensing Strategy · p. 2
  • A saddle-mounted piezoelectric transducer detects rider use and triggers node wake-up when the rider sits down.Otherwise, the remaining sensors can stay disabled so the platform remains in a low-power state.Source: B. Hardware and Sensing Strategy · p. 2

C. Low Power Design Considerations

SmartBike uses deep sleep and event-driven wake-up to limit activity while parked, then adapts sensing and communication behavior to its application. Low-power GNSS operation and transmission of processed state variables further reduce unnecessary energy use.

C. Low Power Design Considerations

  • The saddle-mounted piezoelectric transducer generates a voltage pulse when the rider sits, waking the microcontroller from a very low-power parked state.IMU-based inactivity detection triggers a return to deep sleep when the bicycle is no longer in use.Source: C. Low Power Design Considerations · p. 2
  • ToF sensors avoid their high-activity operating level, and GNSS uses low-power AlwaysLocate mode to adapt fix behavior to motion conditions.The design also adapts sensing and communication behavior to the application.Source: C. Low Power Design Considerations · pp. 2–3
  • The node transmits processed state variables instead of raw sensor streams, reducing payload size and airtime.The design coordinates sensing, embedded processing, and wireless communication rather than maximizing activity in every subsystem.Source: C. Low Power Design Considerations · p. 3

III. IMPLEMENTATION

The SmartBike hardware is integrated around a custom PCB, with sensors positioned for blind-spot coverage and a protected, low-profile installation on the bicycle.

III. IMPLEMENTATION

  • A custom PCB integrates the node’s sensors and power-management system.The time-of-flight sensors cover left, rear, and right blind-spot regions and are housed to minimize interference.Source: §III · p. 3
  • The 3D-printed case houses the IMU, PCB, and battery and protects them from rain and sun.The piezo-electric transducer sits beneath the saddle, while the GNSS module is positioned for sky visibility.Source: §III · p. 3
  • Fig. 4 presents the architecture of the SmartBike firmware state machine.Source: Figure 4 · p. 3

A. Embedded Firmware and Edge Processing

SmartBike firmware uses a five-state Zephyr RTOS architecture and decoupled, non-blocking sensing and communication. Its 20 Hz inertial pipeline estimates roll, removes gravity, and integrates acceleration into velocity with drift mitigation.

A. Embedded Firmware and Edge Processing

  • The Zephyr firmware organizes execution into initialization, sampling, processing, transmitting, and sleeping states.Non-blocking sensor reads and queue-based exchange decouple sensing, processing, and communication; GNSS parsing uses interrupt-driven UART and a ring buffer.Source: A. Embedded Firmware and Edge Processing · p. 3
  • The IMU-based velocity estimator runs in a separate low-priority thread at 20 Hz and processes offset/noise, roll, gravity, and velocity in four stages.Source: A. Embedded Firmware and Edge Processing · p. 3
  • The estimator corrects sensor offsets, applies a moving-average filter, and refines accelerometer roll with gyroscope data.The fused roll estimate supports gravity removal from measured acceleration before velocity estimation.Source: A. Embedded Firmware and Edge Processing · pp. 3–4
  • Trapezoidal integration of processed acceleration estimates bicycle velocity.High-pass filtering and standstill-triggered integrator resets reduce integration drift.Source: A. Embedded Firmware and Edge Processing · p. 4

B. Wireless Communication and Gateway Integration

SmartBike broadcasts compact BLE advertisements instead of maintaining connections, while the gateway scans, validates, decodes, and forwards packets for real-time visualization. This division keeps the bicycle node lightweight and makes communication part of the end-to-end monitoring pipeline.

B. Wireless Communication and Gateway Integration

  • The processed-state BLE payload is approximately 11 bytes, encoding threshold-based blind-spot alert bits rather than full sensor distances.Scaled integer coordinates, estimated velocity, alert flags, and status indicators are included in the compact packet.Source: B. Wireless Communication and Gateway… · p. 4
  • The gateway filters advertisements by manufacturer identifier, checks payload length and structure, decodes values, and forwards them to a local backend.The backend timestamps and organizes incoming data for the real-time dashboard.Source: B. Wireless Communication and Gateway… · p. 4
  • The edge-to-gateway pipeline uses wireless communication as both an energy-aware data path and the link to the monitoring interface.The gateway handles packet reception, aggregation, and visualization, keeping those tasks off the bicycle node.Source: B. Wireless Communication and Gateway… · p. 4

IV. RESULTS, CHALLENGES, AND FUTURE WORK

SmartBike integrates cyclist-awareness functions with compact BLE advertisements and gateway monitoring on a battery-powered bicycle platform. The evaluation estimates long commuting life, finds sensing dominates energy use, and identifies prototype limitations and future improvements.

IV. RESULTS, CHALLENGES, AND FUTURE WORK

  • Compact BLE advertisements transmit processed state information reliably without persistent connections or complex synchronization.The communication layer remains lightweight in the end-to-end prototype.Source: §IV · p. 4
  • Power measurements show that higher IMU operating rates increase consumption, while ToF sensors are among the dominant contributors during active use.Sleep-mode power remains low, and active operation adds limited overhead.Source: §IV · p. 4
  • With a 12600 mWh battery, estimated lifetime is approximately 21 months at 40 minutes per day and 6.5 months at 200 minutes per day; sensing exceeds 90% of energy use, while BLE is minor.The energy summary is provided in Table IV.Source: §IV · p. 4, Table IV · p. 5
  • Piezo-triggered wake-up and application-specific sensing rates reduce unnecessary energy use, while gateway offloading keeps decoding and visualization off the bicycle node.The bike node retains safety sensing and lightweight inference; the gateway handles packet decoding, aggregation, and visualization.Source: §IV · p. 5
  • The prototype used an external battery bank because devboard powering limitations prevented validation of the intended external-powering and harvesting paths.The planned coded-PHY, external-antenna long-range setup was also replaced by standard BLE 1M PHY and the on-board antenna.Source: §IV · p. 5
  • Future work proposes BLE coded PHY and improved antenna integration, more event-driven sensing and selective GNSS activation, plus storage, analytics, and multi-bike support.These changes are proposed to extend range, improve energy efficiency, and broaden the platform beyond a proof-of-concept demonstrator.Source: §IV · p. 5

V. CONCLUSION

SmartBike combines cyclist-awareness functions with event-driven activation, application-specific sensing, local data reduction, and compact BLE broadcasting. The estimated lifetime is approximately 21 months under typical commuting conditions, although external powering and long-range BLE remain limitations.

V. CONCLUSION

  • Piezo-triggered wake-up, application-specific sensing rates, local data reduction, and compact BLE broadcasting reduce unnecessary energy use while preserving real-time operation.The platform combines blind-spot monitoring, rider usage detection, localization, and gateway-assisted monitoring.Source: §V · p. 5
  • Approximately 21 months of lifetime is estimated with a 12600 mWh battery under typical commuting conditions of 40 minutes per day.The paper attributes the estimate to an energy-aware design combining sensing, processing, and wireless communication.Source: §V · p. 5
  • External powering and long-range BLE support remain limitations of the demonstrated architecture.The paper concludes that the energy-aware SmartBike architecture is feasible for sustainable urban mobility.Source: §V · p. 5
Loading 2610.11824v1…