Published January 28, 2023

Asset Management in Remote Location

Asset Management and theft detection in remote location using edge impulse and blues wireless

IntermediateFull instructions provided3 hours339

Things used in this project

Hardware components

Raspberry Pi 4 Model B

Blues Notecard (Cellular)

Blues Notecarrier-Pi

Antenna, Cellular / LTE

Espressif ESP32 Development Board - Developer Edition

DFRobot 6 DOF Sensor - MPU6050

Software apps and online services

Arduino IDE

Edge Impulse Studio

Hand tools and fabrication machines

3D Printer (generic)

Story

Introduction

Remote storage places often faces missing, misplacement and theft of goods in the place. One good example is cell tower backup battery theft in remote areas. Most of the times authorities were alerted on the situation whether it is a theft or misplacement only after too much time passed. Having a high bandwidth continuous streaming solution is not a practical one in these situation.

The solution here tracks each asset in the storage and detect any human presence. Once it detect any presence or movement of any asset it will immediately send alert to the authorities.

Overall Setup :

BLE beacon device is made with esp32 and mpu4050. Using this device we train the data for "MOVING", "FALLING", "IDLE" and "TAMPERING". We will use these trained data in arduino to send different advertisement data based on the state.

EDGE IMPULSE : Data Acquisition and Training

Edge impulse command line is installed in a ubuntu sytem.

refer this : https://github.com/edgeimpulse/ei-install-scripts

The setup

Create a project in edge impulse.

Connect the sensor device to PC.

Upload the following code to get comma seperated date from device.

#include <Adafruit_MPU6050.h>
#include <Adafruit_Sensor.h>
#include <Wire.h>
#define FREQUENCY_HZ        60
#define INTERVAL_MS         (1000 / (FREQUENCY_HZ + 1))

Adafruit_MPU6050 mpu;
static unsigned long last_interval_ms = 0;
void setup() {
Serial.begin(115200);
Wire.setPins(19,21);

if (!mpu.begin()) {
Serial.println("Fail MPU6050 chip");
while (1) {
delay(10);
}
}
Serial.println("MPU6050 Found!");
mpu.setAccelerometerRange(MPU6050_RANGE_8_G);
Serial.print("Accelerometer range set to: ");
switch (mpu.getAccelerometerRange()) {
case MPU6050_RANGE_2_G:
Serial.println("+-2G");
break;
case MPU6050_RANGE_4_G:
Serial.println("+-4G");
break;
case MPU6050_RANGE_8_G:
Serial.println("+-8G");
break;
case MPU6050_RANGE_16_G:
Serial.println("+-16G");
break;
}
mpu.setGyroRange(MPU6050_RANGE_500_DEG);
Serial.print("Gyro range set to: ");
switch (mpu.getGyroRange()) {
case MPU6050_RANGE_250_DEG:
Serial.println("+- 250 deg/s");
break;
case MPU6050_RANGE_500_DEG:
Serial.println("+- 500 deg/s");
break;
case MPU6050_RANGE_1000_DEG:
Serial.println("+- 1000 deg/s");
break;
case MPU6050_RANGE_2000_DEG:
Serial.println("+- 2000 deg/s");
break;
}
mpu.setFilterBandwidth(MPU6050_BAND_21_HZ);
Serial.print("Filter bandwidth set to: ");
switch (mpu.getFilterBandwidth()) {
case MPU6050_BAND_260_HZ:
Serial.println("260 Hz");
break;
case MPU6050_BAND_184_HZ:
Serial.println("184 Hz");
break;
case MPU6050_BAND_94_HZ:
Serial.println("94 Hz");
break;
case MPU6050_BAND_44_HZ:
Serial.println("44 Hz");
break;
case MPU6050_BAND_21_HZ:
Serial.println("21 Hz");
break;
case MPU6050_BAND_10_HZ:
Serial.println("10 Hz");
break;
case MPU6050_BAND_5_HZ:
Serial.println("5 Hz");
break;
}
Serial.println("");
delay(100);
}
void loop() {
sensors_event_t a, g, temp;
if (millis() > last_interval_ms + INTERVAL_MS) {
last_interval_ms = millis();
mpu.getEvent(&a, &g, &temp);
Serial.print(a.acceleration.x);
Serial.print(",");
Serial.print(a.acceleration.y);
Serial.print(",");
Serial.println(a.acceleration.z);
}
}

On the command line type:

edge-impulse-data-forwarder

it will prompt sign in and to choose project.

Device will apper in edge impulse.

On the Data aquisition tab add label and get sample data from device.

Train the data:

Download arduino code from edge impulse and add as library.

Rewrite the esp32 with new code for beacon.

Raspberry PI:

Once OS is boot upload the python code to RPi and Run it

Data will appear in cloud:

Schematics

Code

#include <Adafruit_MPU6050.h>
#include <Adafruit_Sensor.h>
#include <Wire.h>
#include <Asset_Management_inferencing.h>
#include <Arduino.h>
#include <ArduinoBLE.h>

uint8_t manufacturerData[7]={0x01,0x02,0x03,0x04,0x05,0x07};
uint8_t current_state = 1;
uint8_t prev_state = 0;
bool dataset = 1;

#define FREQUENCY_HZ        60
#define INTERVAL_MS         (1000 / (FREQUENCY_HZ + 1))
#define RED 16
#define GREEN 17
#define BLUE 18

// objeto da classe Adafruit_MPU6050
Adafruit_MPU6050 mpu;

float features[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE];
size_t feature_ix = 0;

static unsigned long last_interval_ms = 0;


void setup() {
  Serial.begin(115200);

  if (!BLE.begin())
    {
        Serial.println("starting BLE failed!");
        while (1)
            ;
    }

  BLE.setLocalName("Asset1");

  Wire.setPins(19,21);
  if (!mpu.begin()) {
    Serial.println("Failed to find MPU6050 chip");
    while (1) {
      delay(10);
    }
  }
  Serial.println("MPU6050 Found!");

  mpu.setAccelerometerRange(MPU6050_RANGE_8_G);
  Serial.print("Accelerometer range set to: ");
  switch (mpu.getAccelerometerRange()) {
    case MPU6050_RANGE_2_G:
      Serial.println("+-2G");
      break;
    case MPU6050_RANGE_4_G:
      Serial.println("+-4G");
      break;
    case MPU6050_RANGE_8_G:
      Serial.println("+-8G");
      break;
    case MPU6050_RANGE_16_G:
      Serial.println("+-16G");
      break;
  }
  mpu.setGyroRange(MPU6050_RANGE_500_DEG);
  Serial.print("Gyro range set to: ");
  switch (mpu.getGyroRange()) {
    case MPU6050_RANGE_250_DEG:
      Serial.println("+- 250 deg/s");
      break;
    case MPU6050_RANGE_500_DEG:
      Serial.println("+- 500 deg/s");
      break;
    case MPU6050_RANGE_1000_DEG:
      Serial.println("+- 1000 deg/s");
      break;
    case MPU6050_RANGE_2000_DEG:
      Serial.println("+- 2000 deg/s");
      break;
  }

  mpu.setFilterBandwidth(MPU6050_BAND_21_HZ);
  Serial.print("Filter bandwidth set to: ");
  switch (mpu.getFilterBandwidth()) {
    case MPU6050_BAND_260_HZ:
      Serial.println("260 Hz");
      break;
    case MPU6050_BAND_184_HZ:
      Serial.println("184 Hz");
      break;
    case MPU6050_BAND_94_HZ:
      Serial.println("94 Hz");
      break;
    case MPU6050_BAND_44_HZ:
      Serial.println("44 Hz");
      break;
    case MPU6050_BAND_21_HZ:
      Serial.println("21 Hz");
      break;
    case MPU6050_BAND_10_HZ:
      Serial.println("10 Hz");
      break;
    case MPU6050_BAND_5_HZ:
      Serial.println("5 Hz");
      break;
  }

  Serial.println("");
  delay(100);

  Serial.print("Features: ");
  Serial.println(EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE);
  Serial.print("Label count: ");
  Serial.println(EI_CLASSIFIER_LABEL_COUNT);

}

void loop() {
  sensors_event_t a, g, temp;

  if (millis() > last_interval_ms + INTERVAL_MS) {
    last_interval_ms = millis();

    mpu.getEvent(&a, &g, &temp);

    features[feature_ix++] = a.acceleration.x;
    features[feature_ix++] = a.acceleration.y;
    features[feature_ix++] = a.acceleration.z;

    if (feature_ix == EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE) {
      //Serial.println("Running the inference...");
      signal_t signal;
      ei_impulse_result_t result;
      int err = numpy::signal_from_buffer(features, EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE, &signal);
      if (err != 0) {
        ei_printf("Failed to create signal from buffer (%d)\n", err);
        return;
      }

      EI_IMPULSE_ERROR res = run_classifier(&signal, &result, true);

      if (res != 0) return;

      /*ei_printf("Predictions ");
      ei_printf("(DSP: %d ms., Classification: %d ms.)",
                result.timing.dsp, result.timing.classification);
      ei_printf(": \n");*/
      Serial.println("Predictions");
      Serial.printf("(DSP: %d ms., Classification: %d ms.)",
                result.timing.dsp, result.timing.classification);
      for (size_t ix = 0; ix < EI_CLASSIFIER_LABEL_COUNT; ix++) {
        //ei_printf("    %s: %.5f\n", result.classification[ix].label, result.classification[ix].value);
        Serial.printf("%s: %.5f\n", result.classification[ix].label, result.classification[ix].value);
        if (result.classification[ix].value > 0.6) {
          if (result.classification[ix].label == "Falling")
          {
              Serial.println("Asset is Falling");
              current_state = 2;

          } else if (result.classification[ix].label == "Movement")
          {
               Serial.println("Asset is Moving");
               current_state = 3;
          } else if (result.classification[ix].label == "Tampering")
          {
            Serial.println("Asset is Tampering");
            current_state =4;
          } else
          {
               Serial.println("Asset is in IDLE");
               current_state = 5;
          }
        }
      }
      feature_ix = 0;
    }

  }
  
  if (prev_state != current_state){
    prev_state = current_state;
    dataset = 0;
    BLE.stopAdvertise();
    BLEAdvertisingData advData;
    advData.setFlags(BLEFlagsBREDRNotSupported | BLEFlagsGeneralDiscoverable);
    advData.setFlags(0x04);
    manufacturerData[5]=current_state;
    advData.setManufacturerData(manufacturerData,7);
    BLE.setAdvertisingData(advData);
    BLE.advertise();

  }

}

void ei_printf(const char *format, ...) {
  static char print_buf[1024] = { 0 };

  va_list args;
  va_start(args, format);
  int r = vsnprintf(print_buf, sizeof(print_buf), format, args);
  va_end(args);

  if (r > 0) {
    Serial.write(print_buf);
  }
}

from bluepy.btle import Scanner, DefaultDelegate
import json
import notecard
import notecard_pseudo_sensor
from periphery import I2C
import time
import threading

current_state = "1"
prev_state = "0"
productUID = "com.gmail.thareeq123:asset_management"
port = I2C("/dev/i2c-1")
card = notecard.OpenI2C(port, 0, 0)
req = {"req": "hub.set"}
req["product"] = productUID
req["mode"] = "continuous"
print(json.dumps(req))
rsp = card.Transaction(req)
print(rsp)

class ScanDelegate(DefaultDelegate):
	
    def __init__(self):
        DefaultDelegate.__init__(self)

    def handleDiscovery(self, dev, isNewDev, isNewData):
        #if isNewDev:
          #  print("Discovered device", dev.addr)
        #elif isNewData:
           # print("Received new data from", dev.addr)
        for (adtype, desc, value) in dev.getScanData():
          if(desc=="Manufacturer"):
            data = value
          if(desc=="Complete Local Name" and value=="Asset1"):
            print("  %s = %s" % (desc, value))
            print(f"MAnufactureer data {data} and {type(data)}")
            global current_state
            if data[11]=="5":
              print("Device is IDLE state")
              current_state = "IDLE"
            elif data[11]=="3":
              print("Device is Moving")
              current_state = "Moving"
            elif data[11]=="2":
              print("Device is Falling")
              current_state = "Falling" 
            elif data[11]=="4":
              print("Device is Tampering")
              current_state = "Tampered"

scanner = Scanner().withDelegate(ScanDelegate())
def scanning_func():
	while True:
		devices = scanner.scan()
		
x = threading.Thread(target=scanning_func)
x.start()

while True:
	if prev_state != current_state:
		prev_state=current_state
		req = {"req": "note.add"}
		req["file"] = "sensors.qo"
		req["sync"] = True
		req["body"] = { "Package": "Asset1", "State": current_state }
		rsp = card.Transaction(req)
		print(rsp)
		print("Data send to cloud")

Credits

Thareeq Anwar k

6 projects • 5 followers

Asset Management in Remote Location

Things used in this project

Hardware components

Software apps and online services

Hand tools and fabrication machines

Story

Introduction

EDGE IMPULSE : Data Acquisition and Training

Raspberry PI:

Custom parts and enclosures

RPI Case

Schematics

Schematic _ Arduino

Code

Arduino Code

Python Code in RPi

Credits

Thareeq Anwar k

Comments

Embed the widget on your own site

Asset Management in Remote Location

Asset Management in Remote Location

Things used in this project

Hardware components

Software apps and online services

Hand tools and fabrication machines

Story

Introduction

EDGE IMPULSE : Data Acquisition and Training

Raspberry PI:

Custom parts and enclosures

RPI Case

Schematics

Schematic _ Arduino

Code

Arduino Code

Python Code in RPi

Credits

Thareeq Anwar k

Comments

Related channels and tags