PRE2017 3 Groep11

From Control Systems Technology Group
Jump to navigation Jump to search

PRE2017_3_Groep11_drafts

Coaching Questions

Coaching Questions Group 11

Problem statement and objectives

Problem statement

The usage of drones has become popular in various fields of work and also for recreational use. Drones have a promising perspective in a lot of different areas, like military, surveillance and package delivery. For this project, a problem for drone delivery will be solved.
Currently, when someone wants to let a drone deliver their parcel, the buyer needs to place a landing pad in a place with enough free space around it, for example in their garden. This is, however, not a good solution, because this landing pad is just a piece of plastic on the ground. In other words, the parcel isn't protected in any way against theft or bad weather. Therefore, there is a need of a better solution, that prevents the package of being stolen, blown away or being damaged because of rain, for example. The aim of this project is to design a new platform that does protect a package from theft or bad weather.

Objectives

  • Develop a way in which a drone can safely drop off a package, considering the following components:
    • The parcel itself, which is carried by the drone
    • The type of drone that can carry a parcel and drop it off
    • The landing platform for the drone at the delivery address
  • Consider the risk of theft
  • Consider the location of the landing platform
  • Consider safety risks for bystanders

Who are the stakeholders and what do they require

Users


Delivery services
A user group of the new product, will be delivery services. They can use the delivery box and the platform as a faster and cheaper way to bring their packages to the customers. A problem of this is that not every country allows drones to fly everywhere, because of privacy. Because of this, the delivery services have to make sure and prove to the government that they do not save imagery. A benefit from the new technology is that the company can deliver packages way faster, namely in less than an hour. This is also very useful for customers because, if they need something quickly, they will no longer have to leave the house. As delivery services become more useful for customers, more people will use the delivery services and the delivery companies will grow. Another benefit for the delivery services is that by using a drone the delivery process will be much cheaper than when the company has to hire a driver. Another thing that makes the delivery cheaper is that the delivery time is way shorter. That means that more packages can be delivered in a shorter time. In short, more customers and faster delivery times ensure a huge growth for the delivery services.

Package buyers (consumers)
Consumers require a delivery method that does not require them to be home at the time of delivery. They need to be able to get a delivery box for an affordable price. The box has to be easily to access, must be thief-proof, and should not pose danger to nearby people. The delivery has to be fast and reliable, while also keeping the products private.

Society


Governments in urban planning

Governments can have a huge impact on the success and eventual implementation of the delivery drones. Governments are in the position to make future neighborhoods, that still have to be build, delivery drone friendly, and change current neighborhoods to be delivery drone friendly as well. To be delivery drone friendly the government can implement designated landing spots for delivery drones, that are easily accessible for the residents. There are multiple ways to add this functionality and different options may be chosen for different neighborhoods. Possible options are: Make a central point for all the drones to delivery its package, make landing spots between 2 houses as a shared depot or give every buyer of a future house the option to implement a landing spot in the roof. As can be seen the governmental influence in this new technology can be quite substantial. For the privacy issue explained in the previous section, the place of the landing pad can have a huge impact on how invasive residents find the delivery drones. For the government it is easier to implement this new function in all the newly build homes and neighborhoods without being invasive, then it is for individuals. E.g. the government can build more easily in the front side of the house, the place where people find the delivery drone technology less invasive. For the government itself, focusing on this technology in urban planning there is much profit to be made in the happiness of the residents. There are less delivery people on the street, which is good for the environment and helps to keep streets in living areas quiet and safe. The governments stimulates the economy, because more people are able to quickly order products from enterprises. Putting time in this system during urban planning can increase the feeling of privacy the residents in urban areas have.

Regulation drones
Privacy was mentioned in the part of consumers/buyers as an important issue against the use of drones. The government can enforce laws for delivery drones services to make sure that the delivery drones do not safe any visual input needed to move around. People would become reassured that the delivery drones that flies over their back garden doesn't make any recording of what they see, and thus safeguarding privacy. The government has enfore other laws that the delivery drones are easily distinguishable from reacreational drones. This way people now what is flying over their property. One last point where the urban planning of the government is important is the regulation of the air traffic of the drones. Drone delivery as a new technology suddenly fill the air with a new form of transport that wasn't used in urban areas before. The government has to regulate the air traffic of the drones with clear laws, this to prevent annoyance of residents towards the drone delivery infratructure and system.

Enterprise


Delivery box manufacturers
The manufacturers want to make a product that is cheap to make, so they can sell it for less and eventually sell more. The production of the designed delivery box isn’t a really cheap production. An advantage of this product is that it reaches a large target group, namely anyone who uses delivery services to order products. Because the target group for this product is so high, it is relatively easy to achieve a high turnover. And with a high turnover, no very high profit margin is needed, so that the product can be sold even cheaper. Another benefit from this delivery box is that it is applicable on any living situation, so every delivery box can be made exactly the same. The fact that the target group is large and the fact that the product can be used in every living situation ensures that mass production will be possible for this product, so it can be manufactured cheap and thus can it be sold relatively cheap.

Delivery box installers
The platforms where the package will be brought to has to be built in or around the place where the customer lives. For some customers this is easy, for example people with a large garden or a large balcony. But there are a lot of living situations, like flats or terraced houses where the customer can’t build the platform on his house himself. A benefit from this is that many people need to do a small renovation to their house, to place the platform. Many construction workers can , with a little extra training, learn how to install platforms on houses. With the ability to place platforms on houses, the target group of those companies gets a lot bigger, because a lot of people need them for the new technology. This means that there will be more work for construction workers and thus will those companies grow bigger and have increased sales.

RPC's

Requirements

  • There should be a clearance of 50 cm from the center of the platform to any other object, so there is enough space to let a drone land
  • The package casing must be as weather resistant as the roof has to be
    • The package may not get wet
    • The package can’t be damage by wind, hail, snow and debris flying around
  • A platform must be able to be used by more than one person
  • Requirements of the government “Rijksoverheid” for unlicensed building:
    • Back/front garden
      • Lower than 1 m
      • Maximum area of 2 m²
    • Roof on back side
      • Have a flat roof or,
      • Can’t stick out of the roof more than 1.75 m
      • 0.5 – 1 m of space to between all sides of the roof and the construction

Preferences

  • The platform should be as compact as possible
  • The platform must be placed in such a way that people can easily access it
  • The package should not encounter too much impact
  • The landing platform should be cheap
  • Easy operation
  • Notification should be send when package arrives
  • The platform should be installable in a wide range of environments
  • The platform should have a neutral appearance

Constraints

  • Platform and drone can identify each other, to insure the right package is delivered at the right platform
  • The platform must be able to cummunicate whether there is room for a new package
  • The package casing must have a lock mechanism
  • A drone must be able to land on the platform, or drop a package on/in it without damaging it
  • The drone may not cause harm to people/pets/other obstacles
  • The platform must be able to receive a package of max 2.5 kg and 40 x 40 x 40 cm

Concepts

Interaction platform-parcel

  • The box must be made of waterproof hard plastic. It can be sent back for a certain amount, like a deposit.
  • While approaching the destined platform, a code confirms if the correct platform is being approached.
  • The drone carries a box with two hooks on corners opposite to the side where the box can be opened.
  • Drone carries a box with a lock; the base will lock it in place with an arm / magnet / etc.
  • Drone carries an ordinary box; the base will put a cage around it.
  • The platform itself holds the parcel instead of an external container. The parcel is transfered from the drone to the platform.
  • The platform where the package is deliverd is placed on top of a spring. The platform will be easy to acces for the drone, after the package is placed on top of the platform it will drop.
    Then the top of the protection box is placed on top of the package.
  • Same idea as above, but the spring is replaced with an lift that drops the package down.
  • A platform that forms around the packages when a certain weight is put on it.
  • A hole in the ground that opens when a code is shown. Otherwise it is closed and you're able to walk over it like a normal tile.
  • For a society with both flying/driving drones:
    • The platform is either placed on a pole(Flying drones) or moved to ground level with two arms(Driving drones).
    • The pole it out of the ground(Flying drones) and can move back into the ground so it is accesible for driving drones and the pole isn't as present.

Lock mechanism

  • The side of the box that can be opened is faced down, so potential thieves can't open it.
  • The hooks on the box clamp on the platform and lock, so the box can't be opened anymore.
  • A code lock, of which the code is shared with the recipient via email.
  • Fingerprint sensor
  • RFID with mobile phone
  • Own standard key/code/password
  • Identification of the right platform/drone
  • QR code
  • RFID
  • Radio / IR / other radiation to transfer information
  • Barcode
  • Box opens after estimated delivery time has gone by.

Location landing platform

  • Roof
  • Garden
  • balcony
  • Out of a window(maybe attached to wall in order to keep window closed).
  • Drone can enter specific part of house.
  • Attached to the wall next to the front door.
  • For appartments, a room easily accesible for drones where a certain amount of landing platforms are(Package need to be retrieved in x minutes, so it can be used by other residents).

"MICHAEL system"

(See source:A Multi-Modality Mobility Concept for a Small Package Delivery UAV)

  • Joint landing platforms for all the people in a neighborhood.
  • Flying drones to deliver to this point.
  • Driving drones deliver to front door.
  • Or, People retrieve the package themselves from the landing pad.

Final design

Chosen concept


"Insert picture concept"

Concept

It has been chosen to make a box as a landing platform, in which a drone can drop a parcel. This platform’s housing has a simple box shape. On the top of the box is space for a delivery drone to land. The platform offers identification, so the drone can verify his delivery address, by means of a bar- or QR-code. The housing contains two cylinders, one inside the other, acting as an access to the storage compartment. The outer cylinder is servers a door or lit for the platform, closing the housing. The inner has a cylinder to receive the package from the drone when the housing is opened. Both cylinders turn around the same axis. The cylinders turn simultaneously, but in opposite direction opening or closing the housing. The drone drops the package in the exposed inner cylinder, the cylinders rotate back into closed position and the package drops onto the package holding platform. The platform lowers the package into the storage compartment. The user can collect his package via a separate door, containing a bar- or QR-code as well for identification purposes.

Advantages
  • The system is quite compact
  • The systems has no out / inward moving parts
    • The drone can't be obstructed/blocked because of these parts
  • This keeps the platform compact
  • Can be placed in various environments
  • Central points
  • The design is efficient and secure
  • The system is easily waterproofed
  • The system is anti-theft because of the identification codes and anti-theft casing
  • The system can be shared with multiple people(depending on the location of the system)

Considerations for choosing this concept

The "MICHAEL system" seem to not be a viable concept to use for our landing pad. The designated landing pad that can be used for multiple people is hard to implement in already excisting neigborhoods, since a decent area is needed to work out the concept that may not be available. Then the person has to either walk to the landing spot to get their package, or they have to be home for the driving drone to deliver the package from the landing area to the front door. This takes away from the ease that is strived for when using the delivery drone service, since you have to be home for to obtain the package. Another way is to have your own box where the driving drone delivers the package into, but then it becomes even more expensive then it already is. Futhermore, the reason to have an shared landing spot in a neigborhood completely fails if people need to get their own safe box system as well.

For the location the front side of the roof the best location for the landing pad, taking into account all the stakeholders. The front side of the house is considered less of a personal space as the back side of the house. Therefore placing the box on the front side of the house will safeguard the privacy more than placing the box on the back side of the house. The roof is a better location than placing the landing pad on the roof for multiple reasons. If the landing pad is placed on top of the roof the package can't be stolen easily by thieves. The drone can spot the landing pad more easily on the roof than when it is on top of your roof, and the drone has a shorter vertical moving time. The safety of bystanders is taking into account as well, since there are no direct bystanders like overexciting dogs or kids that can interfere with the system. The only downside is that for already existing homes it will cost extra money to install the system onto the roof and a building license is needed for the front side.

The enterprises want to have a reach that is as big as possible. Taking into account the whishes of the enterprise the system is also usable in the front and back yard, and the back side of the roof. The system works in these locations in the same way, the difference is that it cost less money for people to use the system and a building license isn't needed. However, the system is less optimized in terms of privacy and safety for these locations. Therfore, the help of government in urban planning is asked. With the help of the government people get the chance to build the system into their future house, getting an optimal service.

Adding the protection of the box already before shipping will cause more problems than that it would solve. Exemples of these system are: "The box must be made of waterproof hard plastic. It can be sent back for a certain amount, like a deposit" and "Drone carries a box with a lock; the base will lock it in place with an arm / magnet / etc.". These system put to much of a burden to the enterprises. These enterprises need to invest a lot of money for all these boxes, and it costs an extra flight for every package to return these protecting boxes. Moreover, doubeling the amount of flights would quite possibly annoy consumers. And the extra weight of the protected boxes would either result in a change of drones or a lower maximum weight of the packages. Therefore, these systems aren't chosen for the final concept.

The two systems that fold/put a cage/protection around the package wouldn't be optimal, because the moving parts can easily obstruct the drone while delivering the package. Furthermore, these systems can only be used on the ground or people need to do acrobatic tricks from their window to retrieve the package. Endangering people into these activities isn't helping anyone and putting the system on the ground was a less optimal choice as explained earlier. The "A hole in the ground that opens when a code is shown. Otherwise it is closed and you're able to walk over it like a normal tile" system seems to be a less optimal choice for the same reason.

Considering the identification/lock part of the system the bar- or QR-code are the best options for all the stakeholders. Every package already has either a bar- or QR-code which makes it easier for the enterprises to adapt to our landing platform. If the landing platform would be shared by multiple household than a bar-code, QR-code or a personal password would be safe and easy to use. If the system is used for only one household a physical key is also a possiblity. There aren't big differences between the identification/lock systems, one is not necessarily much better than the other. For the final concept the QR-code has been chosen. The code can be send to you via email, en than you can use it to unlock the box. This way multiple people can use the system, no one needs to remember a password or keep a physical key and the system is still safe from theft.

Prototype

Raspberry Pi

Raspberry Pi Connections
Python Code

I've got a Raspberry Pi 3 model B to control our prototype using python 3. The dimensions are 9x6x3, and connections will be on the sides shown in the figure.

I spent a lot of time learning how it works, but in the end I managed to install a QR-code reader on it by using "pyzbar" (a zbar distribution for python 3, which is a QR-code reader), cv2 (an image reader) and urllib (a url-reader). I temporarily used an old phone as camera, since I didn't have anything better suited. This phone broadcasts its camera with the app "IP camera", which allows the url-reader (urllib) to download images to the Raspberry Pi. cv2 and pyzbar can then read any QR-code in the image, verify it and make the storage box open.

Currently, when the python code is started, an image is read once, and "success" is displayed if the QR-code is succesfully read and correct. The Python code shown on the right (click to enlarge/read). There is still a lot to do:

  1. Determine what kind of camera we should use (Phone is temporary, check if TU/e has camera)
  2. Determine if we want video stream or pictures at certain time interval
  3. Work out how to make script iterative until code is read
  4. Work out how to make box open when code is read (how to control pins to send signal to servos/motors)
  5. Determine when we want the box to close after QR code is succesfully read and box is opened
  6. Make box close after package is dropped.

We still need to decide whether to buy a new webcam, or just use the old phone from Camiel. A webcam would cost about 15-20 euros, so it is also a matter of budget.

PRE2017_3_11_Python_Code

Electrical circuit

"insert schematic electrical circuit"

Materials needed for prototype

Part Amount Costs Store Ordered? Paid By
Raspberry Pi 3 Model B 1 € 0 Borrowed (Camiel) x Camiel
5V 2A Adapter 1 € 8.00 Tinytronics x Camiel
Continous servo 2 € 16.00 Tinytronics x Werner
Breadboard 1 € 2.50 Tinytronics x Werner
Male-Female Jumper Cables 20 € 1.50 Tinytronics x Werner
Male-Male Jumper Cables 20 € 1.50 Tinytronics x Werner
DC Jack female adapter 1 € 1.00 Tinytronics x Camiel
AA Battery 12 € 6.99 Conrad x Camiel
USB camera 1 € 13.49 Conrad x Camiel
Ball Bearings 6 € 14.94 Conrad x Camiel
MDF (laser cut) (4x600x300mm) 4 € 44.35 DigiFab x Jasper
Simple lock 1 € 10.95 Hornbach x Camiel
Hinge 2 € 3.40 Hornbach x Camiel
Dowel (8mm x 100 cm) 1 € 0.50 Hornbach x Camiel
Bolts and Nuts (M4 x 16mm) 30 € 4.58 Gamma x Camiel
Kill Switch 1 € 0 Borrowed (Camiel) x Camiel
Total costs € 129.7

NX drawings prototype

Cylinder gears.jpg
Cylinder.jpg
Cylinder side.jpg
I can't upload my screenshots. an attempt on uploading them will be made to morrow. other wise they will be shown during the tutor meeting.

Approach, milestones and deliverables

Approach

First we will study the literature to figure out to what extent our case is already researched. After that we will attempt to determine the present day problems with the technology. From the present status of the technology we will make a list of RPC's. With that list we will create concepts, work out the best concept and assemble/build the prototype. In the design process it is very probable that we need to go back a step to eventually make a better design. In the design phase we may need to measure the drones that are available in Gemini to determine the final measurements of the design. And in the testing phase we might need drones to check if the system react to a flying drone like we designed.

Milestones

  • Literature study
  • Sources read
  • Sources summarized
  • Conceptualize
  • RPC's
  • Concepts/drawings of concepts
  • Decide a final concept
  • Work out concept
    • Choose/order materials
    • Choose/order operating system/electronic parts(Like arduino was in Engineering Design case)
    • Costs overview
    • Building/assembling prototype
  • Test prototype
  • Conduct a survey about the prototype

Deliverables

  • A technical report of the design process(Wiki)
  • A prototype of the landing pad
  • NX design of the real product
  • Final presentation explaing the design process and showing the prototype

Survey

A setup for a survey has been made. We decided not to hold it right now, since we don't have time to wait until the results come in, because we need to start building the prototype soon. Because of this we decided to hold the survey after building the prototype, when we have pictures and more questions.
https://docs.google.com/forms/d/1iMVSY1XEiBxn46ks75MAHNXn72QkF69rqFd5D89IUjg/edit?usp=sharing

Who's doing what

Week Task Responsible
1 Write introduction Everyone together
Update the wiki Everyone for themselves
Search for sources (10-2)
Summarize sources (19-2)
2 Rewrite subsections made in Week 1 Everyone together
Make a final RPC list out of the personal ones
Combine the concepts in one clear list
Specific literature study (Summerize 2 sources pp) Everyone for themselves
Think of RPC's for the landing platform
Describe or draw concepts
3 Set up a survey Camiel
Rewrite the RPC's Werner
Work out and compare concepts Kobus, Michael and Jasper
4 Finish final Concepts Everyone together
Put the model in NX Kobus and Jasper
USE analysis buyer/customers Camiel
USE analysis companies Werner
USE analysis society/legislation Michael
Determine materials for actual product Michael
Determine materials for prototype Mostly Jasper and kobus, discussed with whole group
Figure out the electronic systems Camiel, Werner and Michael
Update wiki Everyone for themselves
5 Prototype hardware Kobus and Jasper
Put the prototype in Siemens NX and make it ready for laser cutting Jasper and Kobus
Finish USE analysis Werner, Michael and Camiel
Finish materials of actual product Michael
Send mail for materials and machine usage Michael
Order parts(Like arduino or building materials) -
Electronic circuit Michael and Werner
Prototype software Camiel
Update Wiki Everyone for themselves
6 Prototype hardware Kobus and Jasper
Finish electronic circuit Werner and Michael
Finish software Camiel
Assembling hardware/software Everyone together
Make/take survey -
Update Wiki Everyone for themselves
7 Make/take/analyse survey -
Finish Presentation Everyone together
Finish prototype software Everyone together
Finish prototype hardware Everyone together
Presenter -
Finish Wiki Everyone together

Sources and summaries of the general subject

[1]Toksoz, Tuna; Gulkaya, Tutku; and Price, Thomas, "Auto-Selection Of Package Delivery Location Based On Estimated Time Of Delivery", Technical Disclosure Commons, (December 22, 2016)
Drone shipping versus truck delivery in a cross-docking system with multiple fleets and products
Drone delivery models for healthcare
Multi-Agent Path Finding with Payload Transfers and the Package-Exchange Robot-Routing Problem
Parcel delivery in an urban environment using unmanned aerial systems: a vision paper
The Sky’s (Not) the Limit - Influence of Expertise and Privacy Disposition on the Use of Multicopters
Post-Production Analysis Approach for drone delivery fleet
A multi-objective green UAV routing problem
The Vehicle Routing Problem with Drones: Extended Models and Connections
Logistics support for a delivery drone fleet
A cost-optimization model in multi-agent system routing for drone delivery
The regulation of civilian drones impacts on behavioral privacy
Understanding security threats in consumer drones through the lens of discovery quadcopter family
Drone-Aided Healthcare Services for Patients with Chronic diseases in Rural Areas
Can unmanned aerial systems (drones) be used for the routine transport of chemistry, hematology, and coagulation laboratory specimens?
Autonomous Aerial Cargo/Utility system
The economic and operational value of using drones to transport vaccines
Drone transport of microbes in blood and sputum laboratory specimens
Privacy and drones: Unmanned aerial vehicles
Civilian drones, privacy, and the federal-state balance
A cost-benefit analysis of Amazon Prime Air
Optimizing a Drone Network to Deliver Automated External Defibrillators

Sources and summaries on the specific case

Robotic Aerial Vehicle Delivery System and Method
Secure Handling of Unsupervised Package Drop Off at Smarthome
Autonomous and automatic landing system for drones
Landing Pad For Unmanned Aerial Vehicle Delivery
Machine-readable delivery platform for automated package delivery
System and Method for Controlling Drone Delivery or Pick up During a Delivery or Pickup Phase of Drone Operation
A Multi-Modality Mobility Concept for a Small Package Delivery UAV
Delivery platform for unmanned areal vehicles
Drone Operated Delivery Receptacle
Unattended Delivery Drop Box
Building laws