Hey Guys,
I am doing a project called "Vehical Health Monitoring and Diagnostic". My Objective for this project is to give all information about the vehical by measuring various parameter based on this it can predict the present faults and future fault for prescheduling of maintanace. I want to centralize this on a internet server and will be making app to interact this data to user.
I want you all to participate in this project and give your suggetions and help. Please post ideas for doing this project.

 

What is your own knowledge on the subject? Any particular design you got for reference.?

It could become rather complex, rather quickly, let me tell you.

 

Welcome to AAC!
How will this differ from the existing OBDC system used in modern vehicles?
Good luck with predicting future faults. The average garage has enough trouble correctly identifying existing faults .

 

I want to centralize this on a internet server and will be making app to interact this data to user.
I want you all to participate in this project and give your suggetions and help. Please post ideas for doing this project.

I'd start by reverse engineering the current products on the market that do this exact chore. No reason to reinvent the wheel.

 

Welcome to AAC!
How will this differ from the existing OBDC system used in modern vehicles?
Good luck with predicting future faults. The average garage has enough trouble correctly identifying existing faults .

Thank You for taking interest.
But that OBDC system can be used by bigger service center only and that is when our vehicle fail. Also not all of the garrages cannot afford this device. So I want to read that by whatever the embedded system. So we want to develop such a system that even I can see where the fault is or which part is giving bad performance this will also help the garrages to find the fault quickly. Being IOT included also thinking to automatically find nearest garrage and schedule repair by preproviding health report.

 

Welcome to AAC!
How will this differ from the existing OBDC system used in modern vehicles?
Good luck with predicting future faults. The average garage has enough trouble correctly identifying existing faults .

Thats why we would be developing this system so that average garrages can also exact fault location.

 

What is your own knowledge on the subject? Any particular design you got for reference.?

It could become rather complex, rather quickly, let me tell you.

I am also new to this project.
I read something about OBDC. so we can take help of OBDC to access various vehicle parameter.
OBDC works on CAN.
So first I am finding which parameters should be consider to estimate vehicle health.

 

Has anyone seen that commercial......think it was Intel AI Cloud I think. Where they plug the AI terminal into the belly of a heavy airliner. And the computer voice tells them to change capacitor C4.

Can't wait.

 

I'd start by reverse engineering the current products on the market that do this exact chore. No reason to reinvent the wheel.

Thank You for taking Interest.
This product are costly and normal people cannot afford it so we want to develop a product with lowest cost which can be done by making it open source. That is whatever I am doing I will make it available to everyone including tutorial also so anyone can develop this system for himself.

 

I dont want to do this project myself and for me only and want your contribution.Your ideas and help.

 

A few things. I think Onstar has some of the capability. GM computes oil change intervals somehow.
I don't know if someone can guess when the brakes need to be replaced by driving habits. You also need t know when service was performed. I think what's most important is car repair history. e.g. The brake lines rust out on these cars. Environment and weather. e.g. You need to change your wipers. I'm being wierd.

 

Have a look on eBay, creader and Bluetooth obd readers are quite affordable now.
It might be possible or easier to catch the data stream via Bluetooth than develop your own obd reader.
Cheers Geoff

 

https://www.seeedstudio.com/CAN-BUS-Shield-V1.2-p-2256.html
https://www.seeedstudio.com/DB9-to-OBD2-Cable-With-Switch-p-2872.html
http://www.ebay.com/itm/like/221821719820?chn=ps&dispItem=1

 

https://www.seeedstudio.com/CAN-BUS-Shield-V1.2-p-2256.html
https://www.seeedstudio.com/DB9-to-OBD2-Cable-With-Switch-p-2872.html
http://www.ebay.com/itm/like/221821719820?chn=ps&dispItem=1

I have one of the ELM327 modules connected to the torque app in my phone. It works well but is limited in its capabilities. I can diagnose the Check Engine Light and see maybe 100+ different real time variables of the vehicle (RPM, O2 levels, timing, engine load, emissions status, just to name a few). It is a very good tool and excellent value for the money. Using that data (if you know how what it means and keep track of the numbers) you could probably predict a lot of faults. But there are some things that it can't do. OBD is a protocol that all new vehicles are required to have. But it is not the only protocol they can have. Most MFGs have additional, more capable proprietary protocols running in the same space that communicate with their proprietary equipment. Some generic OBD readers can read more than the ELM327 units; My mechanic friend has a $200+ Bosch unit can diagnose my ABS fault, while my $10 ELM327 cannot detect an ABS fault at all. If I took it to the Chevy dealership, their $20,000 unit would probably have a quick discussion with the ABS module over a secret encrypted channel, and convince it to fix itself.

 

What you think what are drawbacks of existing system.
What I think is
(1)User should be connected with OBD2 that is cannot check remotely.
So if I am in vehicle renting business I have some cars and drivers I am not able to see my vehicle health from my home when having a cofee.
(2)It keep this information to mobile app only what I can do is automatically schedule my repairing with nearest and free garrage and Inform him previously about my vehicle fault so he may be ready with solution.
(3)Once I connected this system to internet with gps I can also track my vehicals location and provide emergency service in case of accident(Auto accident detection) by sending location.
(4)I will not have to worry about servicing time it will automatically alert me and give me servicing scheduling option.
(5)
(6)
(7)
(8)what else you think...

 

If I took it to the Chevy dealership, their $20,000 unit would probably have a quick discussion with the ABS module over a secret encrypted channel, and convince it to fix itself.

Funny.

From what I think I know:

The prior unit is the Tech 2 scanner. The new way will be by subscription only. A backyard mechanic with a scanner can rent access for 3 days for like $75.00. Thus ensures that you alway have access to the latest firmware and you need internet access to use it.

Things related to security would be restricted to dealers and locksmiths.

 

Years ago ELM327 was open source. Just because the open source ElmScan source code for the ELM327 isn’t available on line and I have it, here it is. The output was text strings. They weren't obvious but you could sort it out if really motivated. The needed alld42.dll is in the zip folder, put it in C:\windows\system32.
This is the ELM website with data sheets & technical info.
http://www.elmelectronics.com/obdic.html
https://www.elmelectronics.com/wp-content/uploads/2016/07/ELM327DSF.pdf

This is the one I got with USB and is compatible with that source code.USB OBD2 OBDII Code Readers Scan Tools Auto Diagnostic Scanner Car Diagnostic Tool Auto Scan Check Engine Light...
The newest ones are Bluetooth nowProfessional Diagnostic Tool OBD2 OBD-II ELM327 ELM 327 V1.5 Bluetooth Car Diagnostic Interface Scanner Works...
And WiFiELM327 WIFI Wireless ELM327 OBD2 OBDII Auto Diagnostic Scanner Tool Adapter for Smartphone / PC / iOS / iPhone...
All less than $20.00

Info here OBD-II PIDs - Wikipedia, the free encyclopedia

 

(1)User should be connected with OBD2 that is cannot check remotely.
So if I am in vehicle renting business I have some cars and drivers I am not able to see my vehicle health from my home when having a cofee.

OK, so you want to roll your own version of GM's OnStar, Subaru's StarLink, Ford's SYNC and so on.

Let's assume getting the information you need while in the car is easy and that the hard part is getting that info onto the internet.

I say it's easy because very cheap OBD readers and GPS units are readily available. Neither of these are smart enough to totalize information and track maintenance intervals but the OBD would give you enough access to information that should - if you can get at it - give enough information for a remote app to compute what to do. Be aware that not all cars use the same OBD protocols. I'm using the term generically, but there are other systems and they're not compatible. The device you use has to work for the vehicle you have.

One solution that you might implement quickly is to build a smartphone into the car running a custom app to do the monitoring and calculations, and then send the info to the web. I've used my iPhone to talk to my OBD reader but I don't think I can be on the internet at the same time. My OBD reader is wifi, and when I'm connected to its wifi network I cannot also be on cellular. Many of these OBD devices are bluetooth and perhaps this would allow simultaneous connection. You might be able to have the phone switch from the car's network to the internet, but I don't think the OS will allow an app to change the data network without user intervention, which I assume you don't want.

This interim solution (interim because I assume you don't want a cellphone in each car) requires that each vehicle has a paid data service plan. Any solution will require this if you want continuous access to every vehicle. Custom hardware to replace the cellphone cannot solve the problem of access to data service. If you can live with intermittent access, you could have the cars "check in" anytime they are in range of free wifi spots such as your offices, all McDonalds, and so on.

Assuming you can solve that problem of getting data service, then you just need a server to synchronize the data to. Many apps synchronize user data back to their home servers, so I think you can take as given that you can do this without too much trouble. Once the data is on the server, you'll need an app that can read the databases, search for 'events' requiring attention, send notifications and so on. That's a lot of programming but I don't see any obvious hurdles beyond the hard work.

 

I think this project is likely to be hampered by a serious lack of resources. Manufacturer's use dedicated pins on the connector and not all of them use CAN. Besides reverse engineering multiple proprietary protocols you still have the problem of fabrication and distribution of functional boards and maintaining a huge firmware base. Maintaining the huge firmware base is the easy problem, finding someone to do it is the hard problem.

 

"(8)what else you think..."

Acquiring the data quickly and cheaply is not the problem. If you complete this project...you will learn that. You just think that's the problem. But when you understand....it will be no mystery to you.

Your project has been successfully done many times before and in many different forms.
The problem is interpreting and relating the data.......to resolve a particular issue.
This is the problem. This is where the time and cost is. Not the gadget acquiring the data.

When we put the data into the computer......the computer will give the mechanic a action list. An un-skilled mechanic will start doing the actions on that list. Each action takes time and cost money. Some mechanics will argue for each action on the list. After all....the computer said so.

A skilled trusted mechanic will look at that list and preform the needed task to resolve the issue. A good mechanic will save you that time and money.

Profiles on mechanics will save you much more.

This is because we don't build our machines and devices with built in health monitors on individual components. But that is changing.......and sad to say....in the future skill will be un-neccessary.