Start free trial
Trending New Genres Authors
EnglishEnglish
EspañolSpanish
简体中文Chinese
繁體中文Chinese (Traditional)
FrançaisFrench
DeutschGerman
日本語Japanese
PortuguêsPortuguese
ItalianoItalian
한국어Korean
РусскийRussian
NederlandsDutch
العربيةArabic
PolskiPolish
हिन्दीHindi
Tiếng ViệtVietnamese
SvenskaSwedish
ΕλληνικάGreek
TürkçeTurkish
ไทยThai
ČeštinaCzech
RomânăRomanian
MagyarHungarian
УкраїнськаUkrainian
IndonesiaIndonesian
DanskDanish
SuomiFinnish
БългарскиBulgarian
עבריתHebrew
NorskNorwegian
HrvatskiCroatian
CatalàCatalan
SlovenčinaSlovak
LietuviųLithuanian
SlovenščinaSlovenian
СрпскиSerbian
EestiEstonian
LatviešuLatvian
فارسیPersian
മലയാളംMalayalam
தமிழ்Tamil
اردوUrdu
Searching...
SoBrief
Intelligent IoT Projects in 7 Days

Intelligent IoT Projects in 7 Days

Build exciting projects using smart devices
Build connected-device prototypes, from soil-aware watering to voice control and cooperating robots.
by Agus Kurniawan 2017 206 pages
Listen
Immersive
V2.3
Amazon Kindle Audible
Try Full Access for 3 Days
Unlock listening & more!
Continue

Key Takeaways

Treat every IoT project as a loop from sensing to useful action

Intelligence needs a full loop. A smart device is more than a sensor connected to the internet. It measures something in the physical world, turns that reading into data, processes the data, and triggers an action. In the gardening project, a soil sensor reports moisture, a controller evaluates the reading, and a relay can switch on a water pump. The same pattern appears in parking, advertising, and robotics, though the sensors and decisions differ.

Place computation where it fits. A small board such as Arduino can read sensors and control outputs, while a server handles heavier processing. A gateway can translate between devices that use incompatible communication formats. A Raspberry Pi can combine several roles, but it lacks analog input, so an analog sensor may require an additional converter such as the MCP3008.

Analysis

What's useful about this framing is that it makes an IoT project testable: each link in the chain has a job, and failures can be located rather than blamed vaguely on connectivity. It also brings systems engineering into maker projects. A sensor reading that never changes an outcome is data collection, not useful automation. The architecture should remain as simple as the task allows, though. Adding a server or gateway can create more failure points, so distributed components earn their place only when their processing or translation is needed.

Choose sensors and computing locations around the decision you need

Start with the question. The book's projects select sensors according to the condition the system must detect. Soil moisture calls for a moisture sensor; a parking gate might use ultrasonic range readings or magnetic detection; a robot can use flame and distance sensors. The hardware is not interchangeable: each sensor measures a different physical signal, and its output may be analog or digital.

Match the board to the signal. Arduino's analog input can read a soil sensor's changing voltage, while a digital pin can read a simple motion signal from a PIR sensor. The board then needs a suitable path to a server, such as Wi-Fi, Ethernet, HTTP, or MQTT, if remote processing is required. The book's central architecture separates sensing, local control, communication, and heavier server computation, while noting that a compact computer can combine these roles.

Analysis

This is a practical form of requirements engineering: specify the decision first, then select the measurement that can support it. A useful extension is to ask what conditions make a reading misleading. A PIR sensor detects movement, not a person's sustained attention; a camera sees an image, not identity with certainty. Sensor choice therefore encodes assumptions about the world. The book's examples make components approachable, but a real design also needs to account for placement, environmental variation, and how the system behaves when readings are missing or implausible.

Use feedback, not a timer, to make watering respond to soil

Measure, compare, act. The gardening system uses soil moisture as feedback. A proportional-integral-derivative controller, or PID controller, compares the measured value with a desired setpoint and combines three responses: the present error, accumulated error over time, and the rate at which error changes. The book first simulates this behavior in Python, then sketches a system in which Arduino sends readings to a computer and a relay controls a pump.

Treat the demo as a starting point. Its example turns watering on when the controller output is positive and off when it is negative. The author explicitly calls this a rough demonstration, not necessarily a good control strategy. Readings need calibration, and soil moisture is only one of the environmental variables discussed. A Raspberry Pi also needs an analog-to-digital converter to read an analog moisture sensor.

Analysis

The project makes a valuable shift from fixed schedules to closed-loop control: the pump responds to measured conditions rather than assuming that every day is alike. That idea is central to automation in agriculture and manufacturing alike. But a controller's sophistication cannot compensate for a poorly calibrated sensor or an unclear setpoint. Pumps also have physical limits, while the example's simple positive-or-negative rule hides those constraints. A safe prototype should test thresholds and failure behavior before it is trusted with a living garden.

Make parking trustworthy by joining vehicle detection, identity, and space status

Parking is several problems. A management system needs to know when vehicles enter and leave, which vehicle is involved, and where a vacant space exists. Ultrasonic, PIR, or Hall Effect sensors can help detect presence. A camera can capture a plate, and the book demonstrates OpenALPR, an open-source license plate recognition library. In one test, a plate is read with about 95 percent confidence; a German plate is recognized more accurately after the software is told the relevant country format.

Connect detection to operations. A parking application stores vehicle and entry-exit transactions, manages sections, and can publish availability through a website or mobile interface. The book also describes background subtraction: compare a live parking image with a reference image of the empty lot to identify changes that may indicate occupied spaces. Counting total entries alone cannot identify which area is free.

Analysis

The important design insight is that no single detector answers every operational question. Entry counts can estimate capacity, but they do not locate an empty stall; plate recognition can identify a likely text string, but a confidence score is not certainty. Combining independent signals can make the system more useful, provided the software handles disagreement rather than silently treating every detection as fact. Cameras also introduce privacy concerns that simple sensor diagrams do not resolve. A deployable system needs clear limits on image collection and retention.

Model vending as explicit states, and test whether coin evidence is reliable

Transactions need a sequence. The book describes a vending machine as a finite state machine, a model in which the machine moves through defined states in response to inputs. A simplified flow waits for product selection, accepts payment, checks whether the payment covers the price, and dispenses the product. A cancelled purchase should return the money. A central controller such as Arduino or Raspberry Pi coordinates the interface, payment checks, and product mechanism.

Sensors can disagree with reality. The book tests coin weight with a load cell and reports readings below the coins' stated weights: a 1-euro coin reads 6.7 grams rather than 7.5, while a 20-cent coin reads 4.5 rather than 5.74. Optical template matching offers another way to identify coins, but a rotated coin may no longer match the stored image. The database should track products, prices, machine capabilities, and transactions.

Analysis

The state-machine model is valuable because it turns a vague purchase experience into cases that can be tested: valid payment, insufficient payment, cancellation, and dispensing. That same explicitness helps prevent a machine from delivering an item before payment is verified. The coin experiments also expose an important distinction between sensing and authentication. Weight or image resemblance can classify an object, but neither method alone establishes that it is genuine. Reliable payment design needs calibration, error handling, and a clear response when the evidence is ambiguous.

Count viewers to improve ad timing, but do not confuse presence with attention

Signage can measure its audience. A digital signage system displays scheduled text, images, or video, often through a Raspberry Pi connected to a monitor. The book reviews platforms such as Screenly, Xibo, and Concerto, then proposes adding sensors and recording viewer counts alongside the ad and display period. A central service can distribute content to local screens, while local servers keep operating at individual locations.

Choose a detection method deliberately. PIR sensors indicate movement, thermal sensors detect heat patterns, and cameras can use face detection to estimate how many faces appear in an image. These signals can support reports about when people are nearby and help identify busy periods. They do not establish whether someone noticed or understood the advertisement. The design also needs enough network capacity to distribute its content and a database model that connects viewer measurements to ads and timestamps.

Analysis

This project illustrates how measurement can improve decisions, but its metric is only a proxy. A detected face may be turned away, distracted, or passing by; a motion sensor can register someone who never looks at the screen. Treating viewer counts as attention would create false precision. Still, even a modest presence signal can help avoid showing content to an empty hallway or can inform scheduling. The distinction between observable behavior and inferred mental state is essential whenever sensors are used to evaluate people.

A voice command is a pipeline, not a microphone feature

Speech becomes an action in stages. A smart speaker captures audio, detects a wake word, converts speech into text, interprets that text, and sends a command to a device. The ReSpeaker board combines microphones and audio hardware with Linux and a microcontroller, making it possible to experiment with speech recognition and GPIO control. The book's example listens for a wake word, records the following speech, and prints recognized text.

Connect words to outputs. With the Microsoft Bing Speech API, the board sends audio for speech recognition and requires internet access and an API key. A basic automation could examine the resulting text for a command such as switching a lamp on or off, then control a relay or another IoT board. This chain separates audio capture, recognition, interpretation, and actuation, so each part can be tested independently.

Analysis

Breaking speech interaction into stages helps explain why a device can hear a person yet still fail to do what was asked. Recognition errors, ambiguous wording, network interruptions, and unsafe command interpretation are different failure modes. The cloud service in this example also means the system depends on connectivity and sends audio beyond the local board. For low-risk demonstrations, that tradeoff may be acceptable; for household control, users need clear feedback and a way to prevent a misheard phrase from triggering a consequential action.

A firefighter robot must perceive hazards while navigating toward its task

A useful robot combines abilities. The firefighter design brings together a mobile platform, a flame sensor, obstacle detection, and a water-sprinkling mechanism. The book demonstrates a flame sensor connected to Arduino and uses a buzzer when the reading passes an example threshold. For navigation, a robot can receive remote commands over Bluetooth, or it can scan its surroundings with a distance sensor mounted on a servo.

Autonomy is a decision loop. The proposed robot initializes its sensors and motors, moves, checks for obstacles, changes direction when needed, and responds if it detects fire. The fire sensor's threshold must be adjusted through testing. The chapter presents this as a design pattern rather than a fully validated firefighting machine: a real build would need to ensure that detection, movement, and extinguishing work together under actual operating conditions.

Analysis

The project is a vivid lesson in integration: a robot is not autonomous merely because its wheels move without a joystick. It must sense, choose, and act while accounting for obstacles and its objective. The example also makes the difference between a demonstration and a safety-critical system impossible to ignore. A threshold that works beside a test flame may not generalize to smoke, heat, reflections, or distance. The concept is excellent for learning control, but fire response requires rigorous testing and human oversight.

Coordinate robot swarms only after defining the shared objective and communication

Cooperation starts with a goal. Swarm intelligence draws on collective behavior in animal groups and applies related ideas to multiple robots. The book recommends defining the task first, then selecting the robot platform, sensors, communication method, and algorithm to fit it. Examples include robots arranging themselves into a formation and robot soccer, where teammates must coordinate while avoiding collisions.

The network enables coordination. XBee radios can exchange data among robots, and DigiMesh firmware lets compatible modules form a mesh network without a coordinator. The book demonstrates sending and receiving messages, including broadcast communication. Particle Swarm Optimization is presented as one algorithm inspired by flocking and schooling behavior; its sample code searches for strong parameter values. The chosen algorithm must also fit the computing and battery limits of the robot hardware.

Analysis

A swarm is not simply a collection of robots with radios. The shared objective determines which information needs to be exchanged and what counts as success. This prevents a common design mistake: adding a sophisticated algorithm before clarifying the task. The book's examples help distinguish communication infrastructure from cooperative behavior, though a working mesh network alone does not solve negotiation or collision avoidance. In practice, the simplest coordination rule that achieves the objective may be preferable to computationally demanding optimization.

Analysis

This is a project-centered introduction to intelligent IoT, not a single argument built around one definitive theory. Its organizing story is the expansion of a basic maker toolkit across seven settings: a garden, parking facility, vending machine, advertising display, speaker, firefighter robot, and robot team. Across them, the recurring lesson is that intelligence comes from connecting physical measurement to a decision and an action. Arduino and Raspberry Pi are not ends in themselves; they are ways to join sensors, software, communications, databases, and actuators into a working system.

The book is strongest as a map of the design space. It introduces readers to soil sensors and pumps, image recognition, database structure, speech services, GPIO, robot navigation, and mesh communication. Concrete component choices and runnable demonstrations lower the barrier to experimentation. The progression also broadens the reader's view of IoT, from one sensor and actuator to systems involving users, networks, and multiple machines.

Its main limitation is the distance between a demonstration and dependable deployment. Several examples use simplified decision rules, fixed thresholds, or sample images. The gardening controller is explicitly provisional, and the coin measurements show why sensor calibration matters. Similar caution applies to face counts, plate recognition, fire detection, and autonomous movement. A prototype can show that a concept is possible without proving that it is reliable, safe, private, or secure in changing conditions.

The most useful way to apply the book is therefore as an engineering starter kit. For each project, define the decision, identify what evidence supports it, and test how the system behaves when that evidence is noisy or absent. That discipline preserves the book's hands-on spirit while preventing a successful demo from being mistaken for a finished product.

Summary in 30 Seconds
A useful IoT device links a physical measurement to a decision and an action. Choose sensors for the condition being detected, then match boards, communication, and server processing to the task. Feedback can make watering respond to soil moisture, while speech control separates audio capture, recognition, interpretation, and device action. Prototype results need careful testing: a plate-reading example reached about 95 percent confidence, but coin weights and detection thresholds varied. A working demonstration does not establish reliability or safety.
Contains spoilers

Last updated:

Report Issue
Want to read the full book?

Glossary

Proportional-integral-derivative (PID) control

Feedback-based control

A control method that compares a measured value with a desired setpoint and combines three responses: proportional response to current error, integral response to accumulated error, and derivative response to the rate of change. In the gardening example, a program uses soil-moisture readings as feedback and produces an output intended to control a watering pump.

finite state machine (FSM)

Defined transaction states

A model of a system as a set of distinct states and transitions triggered by inputs. The vending-machine example moves from product selection to payment validation and dispensing, with cancellation and refund handled as part of the flow. Making states explicit helps clarify what the machine should do at each stage of a transaction.

swarm intelligence

Collective behavior model

An approach to coordinating agents inspired by the collective behavior of social animals such as ants, birds, wasps, and honey bees. In the book, robots sense their surroundings and communicate with peers to pursue a common objective. The design must connect the objective to suitable hardware, communication, and an algorithm.

Particle Swarm Optimization (PSO)

Population-based search algorithm

An optimization method inspired by flocking birds and schooling fish. Candidate solutions, represented as particles, update their positions using information about their own best result and the best result found by the population. The book presents a Python example that searches for parameters of a mathematical function, and identifies PSO as one possible swarm algorithm.

DigiMesh

XBee mesh networking firmware

Firmware used with compatible XBee radio modules to support mesh networking. The book describes configuring modules with a shared network identity and exchanging messages among devices. In the described setup, the network does not require a coordinator, allowing multiple robot controllers to communicate over the radio network.

background subtraction

Scene-change detection

An image-processing technique that compares a current camera view with a reference background to identify changes. In the parking example, an image of a lot without vehicles serves as the baseline; differences in later images can indicate that vehicles occupy spaces. The method depends on a suitable background image and consistent scene conditions.

template matching

Image-pattern comparison

An image-processing technique that searches a target image for a region resembling a supplied template. The vending-machine example uses a picture of a euro coin as a template and looks for a match in an image containing coins. Its limitation in the example is sensitivity to changes such as coin rotation.

OpenALPR

License plate recognition library

An open-source library for automatic license plate recognition. The book demonstrates running it on a Raspberry Pi to extract candidate plate text from an image and assign confidence scores. Recognition can depend on the expected regional plate format, and the examples show that results may vary between images and configurations.

Download PDF

To save this Intelligent IoT Projects in 7 Days summary for later, download the free PDF. You can print it out, or read offline at your convenience.
Download PDF
File size: 0.33 MB     Pages: 14

Download EPUB

To read this Intelligent IoT Projects in 7 Days summary on your e-reader device or app, download the free EPUB. The .epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.
Download EPUB
File size: 1.43 MB     Pages: 18
Want to read the full book?
Follow
Listen
Home
Swipe
Library
Get App
Try Full Access for 3 Days
Listen, bookmark, and more
Compare Features Free Pro
📖 Read Summaries
Read unlimited summaries. Free users get 3 per month
🎧 Listen to Summaries
Listen to unlimited summaries in 40 languages
—
❤️ Unlimited Bookmarks
Free users are limited to 4
—
📜 Unlimited History
Free users are limited to 4
—
📥 Unlimited Downloads
Free users are limited to 1
—
Risk-Free Timeline
Today: Get Instant Access
Listen to full summaries of 26,000+ books. That's 12,000+ hours of audio!
Day 2: Trial Reminder
We'll send you a notification that your trial is ending soon.
Day 3: Your subscription begins
You'll be charged on Nov 25,
cancel anytime before.
Consume 2.8× More Books
2.8× more books Listening Reading
Our users love us
1M+ readers
Trustpilot Rating
TrustPilot
4.6 Excellent
This site is a total game-changer. I've been flying through book summaries like never before. Highly, highly recommend.
— Dave G
Worth my money and time, and really well made. I've never seen this quality of summaries on other websites. Very helpful!
— Em
Highly recommended!! Fantastic service. Perfect for those that want a little more than a teaser but not all the intricate details of a full audio book.
— Greg M
Save 62%
Yearly
$119.88 $44.99/year/yr
$3.75/mo
Monthly
$9.99/mo
Start a 3-Day Free Trial
3 days free, then $44.99/year. Cancel anytime.
Unlock a world of fiction & nonfiction books
26,000+ books for the price of 2 books
Read any book in 10 minutes
Discover new books like Tinder
Request any book if it's not summarized
Read more books than anyone you know
#1 app for book lovers
Lifelike & immersive summaries
30-day money-back guarantee
Download summaries in EPUBs or PDFs
Cancel anytime in a few clicks

Why one summary?

Scanner
Find a barcode to scan

⭐
⭐
⭐
We have a special gift for you
Open
38% off
DISCOUNT FOR YOU
⭐
⭐
⭐
$79.99
$49.99/year
Only $4.16 per month
Continue
2 taps to start, super easy to cancel
Settings
General
Widget
Loading...