Install Theme

Your web-browser is very outdated, and as such, this website may not display properly. Please consider upgrading to a modern, faster and more secure browser. Click here to do so.

Driven Computing Blog

Welcome to the official blog of Driven Computing.
November 15, 2013

Hacking on Toyotas Quantified Car

On Saturday, November 9, the Toyota ACM Quantified-Car Hackathon kicked off with inspiring words by representatives from Toyota ITC in Mountain View, CA  and Toyota HQ in Japan.  They spoke about how the car is at the beginning of an evolving social system and that cars will see a lot of innovation in the next 5 years. During Saturdays lunch, the Toyota ITC team graciously gave us about 10 minutes to present about the Connected Car Landscape and API’s. 

Toyota is at the beginning stages of developing its API and I appreciate that they are reaching out to the 3rd-party developer community to offer feedback so early in the process.  Here’s a roundup of the 15 projects that were presented at the end of the weekend.

1. Cloud Safety Engine works with raw and augmented data to work in real-time and in historical use cases to create safety-oriented information.  It detects dangerous intersections, road/weather hazards, and tries to predict accidents.

2. Know Me, Know My Car is a contextual in-car productivity/task tool.  The app is sensitive to driving conditions - cruising, stop-and-go, stopped, and active - then bundles tasks based on the driving state.  For example, it allows you to make phone calls while cruising on the highway and was able to pick up context in the phone conversation to create calendar appointments. This team won the prize for Best User Experience in a Car-Relevant Application.

3. Drive Karma offers a driving score in three ways.

  1. Social driving score - likes/dislikes to drivers around you

  2. Car recorded score - uses data from the car to create a score based on how user interacts with car

  3. Eco-driving score

4. Toyota API + ProjectWare (a business intelligence dashboard) flags key events based on data pulled from car.

5. Auto Big Data reports how the car has been driven and used by the driver, like CarFAX. This is one of my favorite projects presented because it solves a real problem.  A few years back, a friend of mine sold his car, which was in a handful of accidents that required repair and body work. When he downloaded the CarFax, there was no record of any of the accidents.  This app offers a more accurate and detailed report using API data in the car.

6. Dead Reckon uses GPS, car speed, yaw data, and acceleration to get very accurate car positioning. They demonstrated how inaccurate standalone GPS systems are by playing a video of a GPS system navigating through downtown San Francisco, where the tall buildings abstract the signal.  The car icon was spinning on the map.  Dead Reckon uses the data from the car to show lane-level accuracy.  They finished their presentation showing the cost-savings: $70,000 LIDAR systems versus their app that used the Toyota API and GPS for less than $50. This team won the prize for the Best Use of the Data API.  

7. Toyota Social is a safety app that detects an accident or hard-braking and then activates the phone camera to take pictures.

8. NoJack notifies you when your car’s location changes while you are away.

9. iDriveSafe compares your driving statistics against the average and makes suggestions when you make driving mistakes, such as sharp turns.

10. Safe Student Driver is an app for concerned parents and teen drivers. It monitors teen drivers and offers a rating.

11. Data Visualization uses vehicle data to find anomalies.  In the case of the hackathon API data, the odometer data jumped from 6k miles to 30k miles.  She looked at the GPS coordinates and found that the data jumped from California to Japan.

12. Reserve a Car Seat finds people going to the same destination/event and algorithmically maximizing spots & routes. I can see this as an employee carpool matching service that companies offer as a benefit.

13. Canary gives status information about the car including location, open doors when you’re not nearby, tow alert, and gas leaks.

14. Car Cost is a mobile app for tracking total cost of ownership for your car via a beautiful interface. This team was very enthusiastic about what they produced and said that they plan to continue building on this idea.  I’m looking forward to what’s next for them.  

15. _Once is a conceptual app of poetic voice commands in the car.

Overall, this was a well-organized event with great energy. The enthusiasm of the developers makes me excited about what Driven is building and how we fit into the Connected Car ecosystem. I’m looking forward to watching the progress of the Toyota API and their community outreach.  

For more photos of the event, please visit Nov. 9, 2013 ACM Toyota Hackathon.

October 17, 2013

The Real Players of Automotive IVI Platforms

There seems to be constant chatter in the telematics sector about the possible threat of consumer software companies such as Apple and Google entering the automotive space.  Telematics Update paper, Can Apple and Google solve platform proliferation for automotive? addressed Apple and Google and whether these companies were capable of creating a de facto platform standard.   Here’s our response as to why these software giants will probably not dominate in the telematics space.

Google 

Google will be involved and influential but will probably not dominate in the automotive sector for several reasons:

  1. Googles core focus is big data and all that relates to it. While there is lots of data produced by a car (Ive heard estimates ranging from 2 to 6 to 16 gigabytes per hour), Google will want to focus on extracting that data and cataloging it.   
  2. Google doesn’t tend to create or maintain industry standards.  If you think of Google as an R&D center, they usually do internal research, develop some prototypes, and then release a white paper with a declaration of what they believe to be the best practices in an area. 
  3. Other companies are the biggest beneficiaries of Android. Android may be an up-and-coming player in the IVI platform world, but it may not be Google who will impact this space.  It may be another company such as Samsung using Android to do it. 
  4. Google might get the most media coverage, but it is not the only autonomous car out there. Carnegie Mellon University has a car that is arguably more advanced.

Apple

Apple has not demonstrated that it’s in the business of wanting to integrate with other systems.  They are often referred to as a walled garden in which they have their own processes and hardware.  This is manageable with standalone personal computing systems. However, the idea of integrating with automotive OEM’s and the many Tier 1 suppliers that make up a car, doesn’t seem like Apple’s way.

QNX

QNX is engineered and designed for embedded systems so it will have that competitive advantage. It’s unix based and they are endorsing standards. QNX will remain a relevant player in automotive.

Tizen

TIzen is a Linux-based open source OS that supports HTML5. The Tizen IVI, the operating system of the Automotive Grade Linux Workgroup, is PC-compatible and will be optimized for a broad set of automotive applications such as the Instrumentation Cluster to IVI. Samsung and Intel are highly-involved with the Tizen project and are engaging with the developer community so we may see a lot of movement from Tizen. 

There should always be multiple players in the IVI platform sector.   This ensures a competitive environment in which innovation and relevant new technology is implemented. Even with a standards compliance, Tier 1 suppliers will always have the pain points around customized development and branding for each OEM. The correlations to the smartphone industry underscores that there are many people who find it difficult to conceptualize the connected car beyond the phone. I ask you to think beyond the phone.

As for third-party developers, they still have the barrier of the cumbersome and time-consuming OEM app approval process. This process is painful because OEM’s will only approve a handful of apps a year and, whenever an update is made to the app, the approval process starts from the beginning. I implore OEM’s to re-evaluate this process because it stifles innovation.

September 18, 2013

Making Connected Cars Accessible at the Ford Electrical Vehicle Hackathon


 
I’ve always loved the idea of contextual computing: the idea that an environmental change can trigger an action and I’ve always wanted to see how this could be applied to a car.  Luckily, I got the chance to try out this idea at the Ford Electrical Vehicle Hackathon last week.  

The team consisted of David Pfeffer, Dave Jensen, Anvitha Jaishankar, Ruchir Patwa, Varun Jain, and me, Liz Slocum Jensen.  Earlier this year, David, Dave, and I founded Driven, a cloud-based API platform for connected cars.  We met the rest of the team at Carnegie Mellon - Silicon Valley a couple of weeks ago when we began sponsoring their student practicum.  We figured that this hackathon would be a good team-building activity where we could hack a car via Ford’s OpenXC platform. What was even more special was that Ford was enabling limited write access to the car for the weekend.

Making the Car Accessible Again

Friday night kicked off with TJ Giuli from the Ford Silicon Valley Research Lab explaining his philosophy for promoting OpenXC to the world. Cars used to be very accessible to the hobby car enthusiasts who wanted to tinker.  However, in recent years, as the systems became more complex and proprietary, the car became closed to the average hobbyist.  In contrast, computers have gone from inaccessible punch card machines to easy interfaces that a child can use.  

Next, Johannes Kristinsson, from Ford, presented information about what’s available on OpenXC and the write access to the car that would be available for the hackathon weekend. Beyond the standard read access, temporary write access included lock/unlock doors, internal climate control, defrost, volume control, and cruise control. I was particularly impressed by having write access to cruise control.  I didn’t have any immediate ideas, but I looked forward to what others would implement.  

Ideation

At 10pm on Friday,  we still had no idea what to build.  I’ve wanted an If-Trigger-Then-Action for cars for a while so I threw it out to the team.  Dave started sketching an architecture and the entire team leaned in and was engaged. Everyone was shouting out fun use cases.  Clearly, this was the idea we would execute. Someone estimated that it would take more than 30 hours to cobble together just the trigger-action part of the project, without even connecting the Ford OpenXC piece. This was a hackathon, we didn’t have the time or resources for that.  We were familiar with Zapier and quickly discovered Zapier’s developer API. At first, I was skeptical about the application of Zapier.  It seemed very developer- and business- oriented.  I started thinking of the use cases based on some of the featured services:

  • When someone posts a new story card in AgileZen, then unlock the car doors.
  • If the car goes outside a geo-fence, then create a pull request in GitHub.

These didn’t seem very practical, but then I started digging deeper:

  • If the car goes inside a geo-fence (e.g. my workplace), then create a TaskRabbit job to get someone to deliver bagels.

This started to seem promising.  There were enough consumer-facing services that could be impressive yet practical enough for hackathon judges.

Teamwork

Dave and David had to learn Zapier’s API from the ground up. They built a web service that implemented REST APIs for actions and REST Hooks to subscribe to triggers. Ruchir, Anvitha, and Varun created an Android app that interfaced with OpenXC to access data streaming from the cars. Meanwhile, I created end use cases that were possible with OpenXC and Zapier.  

There were times when we were all in The Zone and focused on our tasks, but when someone ran into challenges there was always another person willing to help out and pair program.

We were testing right up until judging started at 8pm on Sunday night. We, accidentally, hit the Zapier service about 600 times in 3 hours but Zapier was very responsive via Twitter and email with our questions and issues. For that, we are incredibly grateful.

Here are some examples of Zaps we created for our demo:


  • When the car enters a geo fence, send an SMS.

  • When the fuel level is below a X%, send an email.

  • If the ignition status is on and the car doesn’t move for more than X minutes, then ask to turn off the ignition

Other Projects Presented

There were some really intriguing projects presented:

Co2 Regulator - Did you know that too much CO2 exposure from recycled car cabin air can cause headaches, confusion, and affect not only your mood but your ability to make decisions?  Via an arduino environment sensor, this app monitors the CO2 in the cabin and ensure that fresh air, rather than recycled air, is emitted from the vents.

Back Seat Driver- This was a fun android app that enables  you to monitor the car’s data - such as speed and location - and remotely send messages to another android phone in the car.

Smart Battery- This project looks at your existing route and the altitude changes to figure out the most efficient way to recharge your electric battery on the downhills.  The app maker made a great case for this as a smart energy solution:  he asserted that you could save up to $700 in fuel costs after a few years.  

Stress Route - This is an app that links health data - in this case the trace data of a Bluetooth-enabled heart rate monitor was overlaid on a driving route.  This confirmed my hunch that, even on a Sunday, I-880 and the Dumbarton bridge are stressful to drive.  

Caravan - This leveraged the write access to the cruise control for a caravaning application so that the trailing was going the same speed as the lead car.  

Smart Brake Light - I’ve been waiting many years for someone to develop a customizable sign that mounts in the window of a car so I can thank other drivers for letting me merge or flip them the bird when they wouldn’t. This was such a beautifully creative project that is probably not street legal, but check out the prototype video here.

The ideas presented were impressively innovative and, in most cases, very fun.  We were honored to win Best Overall Project. 

Overall Impressions

Ford Silicon Valley Lab is very interested in how people want to use the data from their vehicles.  On one hand, they need to be conservative about releasing access to car functions, but on the other hand, they are trying to push their comfort level and test accessibility. Ford SVL put together this great event to promote the idea that if you open access to the vehicle as a platform, then some very innovative ideas can be produced. Most importantly, this weekend hackathon enabled open discussion between developers, connected car enthusiasts, and Ford.

July 8, 2013

Infographic: Autonomous Self Driving Cars

We recently found this great infographic all about autonomous self-driving cars. Its full of really interesting statistics and the message is clear, in less than 30 years most cars on the road wil be self-driving, which will make the roads safer and save lives.

Although this infographic explains the safety and environmental benefits of future cars, one aspect it missed is time saved. Not only will autonomous cars solve traffic problems, they will also free up the time of former-drivers while riding in the vehicle.



From: Bankrate Insurance’s InsuranceQuotes.com

Tags: autonomous vehicles connected car

June 26, 2013

An Infographic History of Programming Languages

At Driven Computing were not only about connected cars and telematics, were also developers building a platform for developers. We want to provide the tools to make it easier for you, developers, to build solutions for connected cars.

We thought the developer community might be interested in this infographic from the folks at VeraCode Application Security.

image

Infographic by Veracode Application Security

1 note Tags: developers programming

June 13, 2013

We are Driven

Our team is building a cloud-based platform for developers to make it easier to create applications for connected cars.

The current problem is that each car manufacturer has created its own connected car platform. This means that application developers need to re-implement their solution for 12 vehicle platforms. We want to unlock the gates to each of the walled gardens so that developers can leverage all that is capable of the connected car without having to seek out each API. 

Please provide us with any of your thoughts or feedback on developing for connected so that we can help you create the next killer app for connected cars.

We look forward to you joining us on this journey.

Tags: company about