View previous topic :: View next topic |
Author |
Message |
WanaGo
Joined: 15 Aug 2008 Posts: 111
|
Posted: Fri Nov 27, 2009 6:04 am Post subject: Vehicle Logging |
|
|
Hello
I was approached at work today from a guy in the vehicle management dept, and he is wanting to get a system build for monitoring how vehicles are being treated on site - specifically forklifts etc.
Basically he wants to monitor things like throttle position, brake pressure, hydraulic pressure, RPM, Speed, Accelerometer and gyroscope with potential GPS also - and datalogging onto USB or Flash card, maybe even with Bluetooth or Wifi.
He wants to monitor how vehicles are being treated as he is in charge of getting vehicles fixed etc, and wants to know the reason for some of the damage that is occuring.
I was wondering if anyone has spotted any modular microcontrollers in terms of something like a processor base, with add in options for gyro and accelerometers, usb etc.
I know I could just design something and make it, but he is wanting something fairly soon, that is preferably modular rather than custom for each vehicle.
It is to be used on vehicles that are not complex, dont feature CANbus like the new vehicles. Some vehicle monitoring that I have seen is designed to just plug into the CANbus on the vehciles and they can log everything, but that isn not what we are after.
Just curious if anyone has played with this sort of stuff, and if anyone has seen anything in their travels that may be suitable?
Thanks |
|
Back to top |
|
|
ollopa
Joined: 03 Sep 2007 Posts: 233 Location: California
|
Posted: Fri Nov 27, 2009 7:19 am Post subject: |
|
|
All that and the kitchen sink, and he wants it by yesterday... And it had better be cheap, right? I think your boss dreams big, or he needs to get his head extracted from his behind.
More to the point, I think an ARM development board running embedded Linux is the easiest route to interfacing so many peripherals--especially Wifi, flash, and USB. Look at something like this, maybe http://www.active-robots.com/products/sensors/sparkfun/6dof-v4.shtml |
|
Back to top |
|
|
WanaGo
Joined: 15 Aug 2008 Posts: 111
|
Posted: Fri Nov 27, 2009 8:28 am Post subject: |
|
|
Haha yeah - not my boss, but a boss from a different dept at work.
Not so much a case of wanting it yesterday, however its in need of having something to work with sooner rather than full design and development time of creating something from scratch.
I have seen these ARM processors before but know nothing about them, ill do some reading.
Basically the sensors on the vehicles would be analog inputs, so 5 or 6 inputs would probably be fine I think. A few digital inputs too. GPS is not essential as I am not sure if it would work that well going in and out of buildings anyway. Wifi is optional, but would be nice in order to get the info off the vehicle without having to remove the card. Bluetooth is just another option, and again not essential - def not bluetooth and wireless both. Accelerometers and gyro - I dont know exactly what his requirements are yet, but I think he basically just wants to know if there have been excessive accelerations or sudden stops (hitting stuff).
I was looking at the arduino options, purely due to there being stackable packages that are already designed etc, more so than the bootloader etc. Could be an option there.
I really need to find out more in terms of requirements.
It would be 12v or 24v powered, and I have no idea what the sensor outputs would be. These micros take 3.3v or 5v inputs so I would have to scale them with dividers I assume, in order to read in anything above 5v.
Bit to learn yet, but all very achievable im sure. |
|
Back to top |
|
|
JC
Joined: 15 Dec 2007 Posts: 584 Location: Cleveland, OH
|
Posted: Sat Nov 28, 2009 2:29 pm Post subject: |
|
|
Wow. I think this is a very big project.
I have done a similar project in the past, and will share a few thoughts.
If you have experience with micros, then the sensor input is easy. If not, then there is a learning curve for this part of the project. This is a learning curve for each sensor, by the way.
The first question to ask is if the vehicles have any sensors with outputs you can monitor. Adding a hydraulic pressure sensor to a system can itself be a big undertaking. Adding accelerometers is straightforward.
Monitoring the throttle position is an interesting requirement. The question is what does it control, is there a simple voltage signal already available that is proportional to the throttle control? Adding a physical connection to a mechanical linkage is not trivial. Integrating gyro and acceleromter data into useful data is non-trivial.
GPS? Certainly doable, but why? Just how far can one drive a forklift? Many people do not understand the limitations of GPS. It is truely incredible that you can tell where you are anywhere in the world, but not precisely. The location, (position), varies from reading to reading, even if the GPS receiver is sitting still. The position can typically vary by up to 10 meters. That is a big degree of variation on a loading dock floor. Car systems have the added advantage of knowing that the car is on a road, and integrate the GPS position with the road data base.
Additionally, as you noted, you may have very poor GPS coverage within a building.
I used Bluetooth for my unit's data link. I later changed to using XBee for other projects of this nature. The BT worked, but the range was limited. It would not be a problem if someone walked out to each vehicle with a laptop and uploaded the vehicle's data. I wanted automatic uploading, and having a greater range makes this much easier.
I used an SD card to store the data between uploads. You need to decide if you need this type of real time storage, or are wiring the site for real time data uploads and storage on a PC, and hence skipping the local data storage on the vehicle(s). If you add local storage, then USB for a thumb drive is convienient, but pretty much "leading edge" for an 8-bit micro, with a significant learning curve. SD cards are cheap, and there are more examples to draw from.
Vehicle electrical systems are notoriously noisy. You will want to google "load dump", and look into protecting your power supply, and your sensor inputs.
Got all that done? Now you have LOTS of data, from a fleet of vehicles. How are you going to present the data in a useful manner for a manager to review quickly? The boss won't want to scan long sheets of numbers. Bottom line, you now need to develop your own PC based software to store, retrive, and display the data, in a useful manner. Ideally the data graphs would show the expected normal range of each parameter, and show, (highlight), when the operator/vehicle exceed the "normal range". Then you wnat to know the date and time that the event occurred, so you have some sort of tracking back to an individual operator. Are people assigned a vehicle per shift? Does the nearest person jump on the nearest vehicle to do the task at hand? Are you adding a keypad ID to operate the vehciles, (or RFID, etc.)? Do you know how much resistance the workers will put up if they have to log into the vehicle to use it?
Is it worth collecting the data, displaying the data, and analysing the data, if you can not track back to individual users?
In summary, the mechanical installations can be challenging, the micro and sensor portion is doable, but non-trivial, the data sotrage and RF is non-tirival, and the data display is itself a huge project.
Just a few thoughts as you think about the task at hand...
JC |
|
Back to top |
|
|
WanaGo
Joined: 15 Aug 2008 Posts: 111
|
Posted: Sat Nov 28, 2009 10:34 pm Post subject: |
|
|
Hi JC
Thanks very much for what your wrote, and yes you are very correct.
I have experience with Microcontrollers somewhat - have been playing with them for about a year now and getting better. I am an Automation Engineer on the site, so am use to using PLC's etc.
The site is huge, its a smelter, so there is lots of places for people to drive the vehicles and 'hide' etc. Not just a simple loading bay etc.
From what I gather, they are not wanting this for a policing type system, but are more wanting it to see how people are driving in general. Flooring it, hitting things, going places they shouldnt etc.
Being a smelter, there are places that some vehicles are not allowed to go, but there is proof to suggest these vehicles are going there from debris found on the vehicles etc.
This is just phase 1 of the project, just to see what kind of things can be captured, and if they want to go further and capture more information etc, or go into logging in on the vehicles etc.
In terms of the sensors and what they are, if they have the appropriate outputs etc, I am not sure yet. I will find out more this week.
Thanks
James |
|
Back to top |
|
|
ATIU
Joined: 15 Oct 2004 Posts: 534
|
Posted: Sun Nov 29, 2009 2:58 am Post subject: |
|
|
Rather than trying to log all data, why not log "exceptions" that they want to catch. This means much less data to deal with. |
|
Back to top |
|
|
beananimal
|
Posted: Mon Nov 30, 2009 9:26 pm Post subject: |
|
|
Honestly... Not to be rude, I don't get it. Just because you can collect data and monitor something does not mean that you should collect data and monitor something.
While I am all for using telemetry to solve problems and/or streamline a process or create a historical record of data, this certainly does not appear to be a warrented use of technology from any standpoint.
Why customize sophisticated technology to report on useless data that nobody honestly cares about? Employees abusing forklifts? Place a few video cameras on the warehouse floor for a fraction of the cost and time in conception, development, deployment, monitoring and maintenance.
Need more data? Place a wireless camera on each forklift. I think that if you use some common sense and basic problem solving logic, you will find that collecting telemetry from the forklifts serves no useful purpose in this situation. Furthermore, collecting such data in a reliable manner and putting it in a form that is actually user friendly and informative is going to be a very costly project in both man hours and money.
From an educational or hobby standpoint I have no issue with the concept, from a business standpoint (based on the scenario you offered) it does not make good sense.
Sorry... just my two cents. |
|
Back to top |
|
|
AdrianJ
Joined: 16 Jan 2006 Posts: 2483 Location: Queensland
|
Posted: Tue Dec 01, 2009 4:56 am Post subject: |
|
|
Following on from beananimal's thoughts, maybe its more productive to encourage ( even pay for ) the rev-heads to go racing on the local go-kart track. If there is no such facility, it might still be cheaper to build one... _________________ Adrian Jansen
Computer language is a framework for creativity |
|
Back to top |
|
|
kazi
Joined: 21 Oct 2005 Posts: 19 Location: Cape Town
|
Posted: Tue Mar 09, 2010 8:26 pm Post subject: |
|
|
I know this topic is a few months old, but just to add, there are systems available that can do all these things.
Systems that was developed for the same reason. |
|
Back to top |
|
|
teccs
Joined: 14 Oct 2006 Posts: 69
|
Posted: Sun May 23, 2010 11:37 pm Post subject: Did you get any traction on this project? |
|
|
Hi WanaGo,
Did you go forward with this? What has happened so far? |
|
Back to top |
|
|
ronye
Joined: 01 Jul 2010 Posts: 1
|
Posted: Thu Jul 01, 2010 11:39 am Post subject: |
|
|
I'm working on the same project at the moment. Monitoring grader operator behaviour on operation and logging every occurance of fault.
Building from scratch is either interesting and intimidating (time consuming).
Try datataker for a shortcut. |
|
Back to top |
|
|
pedro
Joined: 24 Apr 2009 Posts: 128
|
Posted: Thu Jul 01, 2010 7:39 pm Post subject: |
|
|
Hello Wanago,
I have a similar board that make what you need. It function is dataloging in real time and save every minute a record on logfile on the sd card, every hour a new file with all the collected data.
Trough the RS485 Bus, is almost possible without limits, to expand the module with active sensors, like a GPS. A optional RF Module can upload the data to a computer. A programm on the PC scan for vehicles that are in reception range and collect the data.
On power up it read from sd card a config file with the vehicle models and al the min and max values for every sensor. If want to monitor a new vehicle model , it only need to add a new record with the values on this txt file.
It can protect in real time the vehicle from serious failure, etc.
It have 4 inputs with optocoupler, various ADC inputs and On/Off outputs
The board can work with 12V or 24V, for Cars, Trucks and Omnibus.
It donīt need a ARM controller with OS. This board works with a Atmega2561
Attached the picture of the board
Regards
Pedro |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You cannot download files in this forum
|
|