Street Smart - 2017 Ixd Thesis @streetsmart-ixdthesis - Tumblr Blog

The future of automotive UX, helping users feel safe with autonomous vehicles.

Most innovation by car makers in autonomous vehicles focuses exclusively on the technology that can make a vehicle drive by itself. But just as important are the interactions inside (and outside) of these vehicles to help passengers feel safe, confident, comfortable, and “in control” when using them.

Vision is an operating system for cars in the age of autonomous vehicles that makes human drivers feel safe on the road with other autonomous vehicles. Unlike standard driving interfaces, it keeps the driver alert with “V2V” communication, visualizing the intentions of other vehicles. This helps human drivers feel better informed and safer about what other cars are doing.

Feature: Traffic Negotiation + Intention Communication

#Thesis Sva-Ixd #feature

“V2V”Communication Experiment (No.2)

You can’t go far on the highway without seeing a semi-truck.

Truckers usually communicate with each other using citizen’s band (CB) radios, but they communicate with you using flashing headlights, turn signals, and trailer lights. As with any communication, context is the key for you to understand why they’re doing it.

The most common signal is used by cars and trucks in oncoming lanes. They will double-flash their headlights when they just passed a police speed trap that you’re heading toward.

More than two consecutive flashes from oncoming traffic signals that there is another type of danger ahead, such as a foreign object on the road and drivers should proceed with caution. These two signals are often confused, but the appropriate action is the same, slow down.

Another signal is for lane change clearance. Because semi-trucks are so long, it can be difficult for the driver to tell if he can safely change lanes in front of you. You can quickly flash your headlights when his trailer clears your car and there is room for him to move back into the driving lane.

What really intrigued me about is the existence of this informal language of road signals—either creative adaptations of the simple communicative tools available on an automobile. My question is —— when we’re communicating with an autonomous vehicle, how do we know these informal languages of road signals was understood by the autonomous vehicle?

My proposal is human driver should know the autonomous vehicle's intentions in the specific scenario (avoid over-communicating). When human driver communicates with an autonomous vehicle through light signals, human driver will receive the visual feedback from AV, better informed by their intentions, build the “trust interaction”.

#Thesis Sva-Ixd #autonomous vehicles #experiment

How to play trick with Autonomous Vehicle? & “V2V”Communication Experiment (No.1)

A car is sitting in the middle of a parking lot has been surrounded by a circle. In the language of road markings, the dotted white lines on the outside say, "Come On In," but the solid white line on the inside says, "Do Not Cross." To the car's built-in cameras, these are indomitable laws of magic for autonomous automobiles.

In this case, the circle being used to arrest an autonomous vehicle—a self-driving car, which relies on machine vision and processing to guide it. By quickly deploying the expected form of road markings—in this case, a No Entry glyph—we can confuse the car's vision system into believing it's surrounded by no entry points, and entrap it

Here is a particular scenario. A blue autonomous vehicle is running on the highway, your car is running behind it. Suddenly, the front truck dropped some obstacles. In this case, by law, the autonomous car is not allowed use the HOV Lane and across the solid line, but the right side has another car so the autonomous car can’t merge to the right lane either. It has to break and stop on the highway!

This is your view in your car when the emergency happens, you can’t see what’s happening in front of the autonomous vehicle. When you find the breaking signal from the autonomous vehicle, it’s too late to slow down.

In this scenario, what if there is a non-autonomous vehicle running in front of you and meet this situation? Might be merging to the HOV Lane and you can avoid the potential car crash or highway congestion.

It makes me think unless we have a dedicated lane for autonomous vehicle or until all the vehicles are replaced by the autonomous vehicle, human drivers have to face the future that we need to share the road with them. But how do I know what types vehicles running surround me? What are autonomous vehicles intentions, can they handle all the traffic situation? How do human drivers set up their expectation?

Through analyzing the cybernetics workflow, when autonomous vehicle detected there are obstacles ahead, central computer quickly analyzing the situation then breaking. Human driver observed the front car breaking signal and slowing down. Since because of the workflow, we need to consider human driver’s reaction time.It will cause the potential car crash.

Based on those questions above, my hypothesis is car recognition and “V2V” intention communication is a way to reduce this concern and correct the expectation from a human driver.

“V2V” communication will better solve the problem, we can easily find the different from the refined cybernetics workflow. When the autonomous vehicle’s central computer decide to break, it will also send the signal to the car behind, human driver’s car will automatically breaking at the same time. Reduce the human driver’s reaction time, in this case, the entire traffic system will coordinate in one disturbance.

Beyond that, human driver will be informed by visual information which overlaid on the windshield and rendered on the dashboard. It is necessary to know what’s happening in their own car to reduce the confusion and concern.

#Thesis Svaixd #autonomous vehicles

Research(How does Autonomous vehicles work?)

Machines are much better at following rules than humans; motorway signs advising drivers to slow down or not change lane to avoid creating jams are often ignored by motorists – not so a computer.

So how do driverless cars work? There are several systems that work in conjunction with each other to control a driverless car.

1.Radar sensors dotted around the car monitor the position of vehicles nearby.

2.Video cameras detect traffic lights, read road signs and keep track of other vehicles, while also looking out for pedestrians and other obstacles.

3.Lidar sensors help to detect the edges of roads and identify lane markings by bouncing pulses of light off the car’s surroundings.

4.Ultrasonic sensors in the wheels can detect the position of curbs and other vehicles when parking.

5.Central computer analyses all of the data from the various sensors to manipulate the steering, acceleration and braking.

What I’ve learnt after research?

I have to admit, in most of the cases, programmed machine is more reliable than human beings. Various advanced radars, cameras empower car able to to “See” more than human driver. Machine Learning heavily trained computer driving like human beings and always following the traffic rules. Technically, all above is about “Learning by Imitation”. All the technologies we’ve applied to autonomous vehicle is to create a “Super Human Driver”, but in nature, still, machine making decisions based on what they ‘Saw’.

But is it always true that what we saw? That is my question. What makes human beings different from machine?

Human makes decision is way more complicated than machine, because lots of internal, external factors affect us, not just right or wrong. We are able to break the rules based on current circumstance, but machine can’t, or at least they’re not allowed.

It gives us some opportunities to play trick with autonomous vehicle.

#Thesis Sva-Ixd #autonomous vehicles

Building the “Trust Interaction”

After defined my value proposition, the challenge that I’m facing is to clear how to deliver the sense of safety to people. In specifically, is what kinds of features they need and how to build the trust interactive in-car human interface that instill confidence, control and a sense of safety when operating or riding in an automated vehicle.

Before I start to build my next prototype, I'm planning to do the couple things listed below:

1. Researching and figuring out current autonomous vehicle’s “Running Standards” and how they dealing with the extreme situations on the road. Identifying the constrains of the current solutions and blindspots.

2. Researching and understanding what’s known as “Trust Interaction” in human interface design.

3. Reading “The Way Finding Handbook -Information Design for Public Places”, need to deeply understanding the design principle for public place, make sure the new car interface be intuitive and informative, at the same time, make sure all the design should has low learning curve, readable and understandable, unifying with current traffic information design standards.

4. Creating the Manual Guide Book for users. Easily drive audience through the scenarios, quickly learn and understand how to interact with the new car interface.

#Thesis Sva-Ixd #autonomous

Value Proposition Refinement:

XXX is an operating system for cars in the age of autonomous vehicles that makes human drivers feel safe on the road with other autonomous vehicles. Unlike standard driving interfaces, XXX helps vehicles and their drivers to make better driving decisions, Instead of merely hiding the data away like recent smart car concepts, it keeps the driver alert by connects cars to each other and visualizes other vehicles intentions, helps human drivers feel better informed and safer about what other cars are doing.

#Thesis Sva-Ixd

Ideas + Criteria Matrix

On the vertical axis, is my User Criteria and Personal Criteria. On the horizontal axis, is my Ideas (just a concise title, not the whole description)

Aim: Check the boxes where they intersect, and see which idea hits the most criteria. Most likely, in the process of doing this I will need to edit some criteria, think of new criteria that are important to me, and/or decide to eliminate some that aren't critical. Also it is possible you will see which idea gets the most "points" but realize in my heart I don't really want to do that one, but another one instead. Anything that helps me narrow down is good.

Value Proposition:

_________, is a car operating system for semi and none autonomous vehicle user. Provide more useful roadway information and concrete analysis.Unlike a traditional driving experience, __________ helps you better make the decision, empowers your vision and the way to communicate with another types of vehicles, thriving for create a more safe driving experience.

Reflections on Alternate Prototype

1.Is it more or less promising? Why?

I think it is too early to say more or less promising to my first round prototype. At this stage, the most important thing or I can say the challenge is keep moving forward, bring the idea to pixel work and show them in front of people. If you have to figure out everything and then execute your plan, you will put yourself in a risk situation, there are lots of unpredictable challenges out there and the time is short.

2.What were unanticipated challenges or successes?

In terms of my first pixel prototype, I think it was successful overall. I was struggling in almost entire semester to find a way to deliver the value in my design area and convince people believe the problem is worth to be solved, that is the thing I think was unanticipated for me, especially when you strong believe in it and think people will believe as well. I was happy finally through my pixel work, people start to understand what I’m working on.

3.What are key insights and implications of this exploration to your thesis exploration and concept?

It is hard to attract people with just some fancy technologies any more. Now, they more care about the logic behind your design. For instance,What research or insights are these design choices informed by? What’s the benefit for the adopter? How to test and prove my concept ? How to simulate particular scenario and drive audience in.

I think for the next step, I’m gonna spend more time to think about what kind of features should I bring to user, what should I delete from my design. Don’t be over-designed and keep those challenges above in mind.

4.Regardless if this is the idea you'll be pursuing moving forward, reflect on the insights learned that may be applied to future prototypes (for example, lessons on reward, cue indicators, etc.).

For the future presentation, I think I should find a new way to better show my car system like how to arrive at this particular design for this display.So that I can put my audience in the particular scene and help them better understand and believe my system is value and benefit for them.

Milestone- Thesis concept statement

My thesis topic is about autonomous vehicle. In specifically, is focusing on designing semi-autonomous vehicle car operating system. Through the heavy researches last semester, I found that when people talk about autonomous technology, the most concern from the potential users is about safety. They mentioned a lot about the way to understand AV's intentions and how to communicate. I think the way of communication between different types of vehicles is a critical design challenge because I believe only let user feel confident and comfortable sharing the road with autonomous vehicles they can embrace this new emerging technology.

In my system, I'll work on building the car recognition, priority negotiation in vehicle to vehicle and vehicle to infrastructure, in car technology with less distraction and also think about what could be done for the journey since the entire roadway system as an ecosystem.

Fresh start brainstorming Part one

These images below are my last semester’s final delivery based on my heavy researches and hypothesis. From the research I found when talking to autonomous vehicle, the most concern that people mentioned was safety and trustfulness. “How to build the trust interaction” is a challenge as a UX designer to help early adopter embrace this new emerging technology.

Getting consumers to trust the technology is just as important as the tech itself !

I strong believe build the communication between different types of vehicles, enhance driver’s isights when they are driving, develop vehicles recognition technology, help driver understanding the autonomous vehicle’s intentions are the ways to improve the feeling about safety. Only people can feel safe, they’re willing to embrace the autonomy industry so that we can better improve this technology.

This car dashboard interface helps human driver recognize there is a full autonomous vehicle running ahead and get informed AV’s attentions in next 5 seconds.

In-car UI rendering the real-time traffic situation through a higher angle than human driver’s. It helps human driver feel more “in control” of the road and confident to make driving decisions.

Traffic negotiation: One important point of building trust with the tangible operation of automated vehicles is that passengers and pedestrians must be able to understand what the automated vehicle system is sensing. For instance, participants have mentioned that when the system includes a visual display of a pedestrian crossing the street, which corresponds to the pedestrian they see from the car reaction, confidence in the system is established. Also, human driver won’t be confused or need to assume autonomous vehicles intentions.

The ability of communicating with roadway infrastructures. Helping coordinate to traffic congestion. City exploration is necessary when encountering road congestion so that a alternate route can be quickly suggested and determined is desired;

#Thesis Sva-Ixd #autonomous vehicles

Hypothesis:

Driver will be distracted by the touchscreen more or less.

Roadway Driver Distraction Test

Unresponsive

Hit the brake

Distraction chain reaction:

Unresponsive + Speeding + Braking + Unstable Emotion

#Thesis Sva-Ixd #assumptions #hypothesies #test

Thesis Assumption Map

Since the advent of the autonomous vehicle, people believe it could be the way to reduce the traffic congestion, improve the roadway efficiency and safety. Obviously, in the next few years, more and more fully and the semi-autonomous vehicle will be running and sharing the road with the driver-operated vehicles.Until all the vehicles can “See” each other on the road, communication between different types of the vehicle still could be the problem. Following the trend and focusing on the near future, my first generation idea is involving the roadway infrastructure as the “eye” for those semi-autonomous vehicles, constantly send the real-time roadway information to them avoid the potential risk. How to add, indicate and layout that real-time road information without distraction and overwhelming need to be considered.

Where could be the medium of those interactions in the semi-autonomous vehicle? Is a touchscreen appropriate way to running the system? Will driver really be distracted or just like they said: “We're the technology friendly generation”.

“Drivers will be distracted by the touchscreen more or less when they are driving” is my hypothesis of my assumption, to help me prove it I started this roadway distraction test.

#Thesis Svaixd #assumptions #hypothesies #test

Thought after feedback

Last week, after talked with Eric about my approach of solving the disconnection between auto vehicle and driver-operated vehicle on the roadway——a car operation system + hardware upgraded infrastructure on the roadway. Eric asked me a question about distraction from the car installed touchscreen which I think it could be the one potential risk of my solution.

My thesis goal is to find a way apply in the near future roadway to help people adapt to the road environment with growing autonomous vehicles and the existing driver-operated vehicles, improve the traffic efficient, congestion mitigated and the most important thing, safety. Because of the proposal, the solution must contribute to the near future and has its advance. Since people have the concern about the distraction from car installed touchscreen, I have to do more researches to figure it out.

#Thesis Sva-Ixd

The near future of smart road

Let’s start by looking at what is perhaps the greatest challenge and most exciting innovation that cities will face in the coming decades: the advent of self-driving and connected vehicles.

I believe that over the next ten years and beyond,autonomous and vehicle-to- vehicle technologies will fundamentally change the way we move.

In terms of V2V, or the dynamic wireless exchange of data between active, moving vehicles, is based on creating a technological ecosystem in which vehicles in close range are in constant communication with one another, greatly enhancing both the safety and efficiency of roadways.

Make the future a little bit closer. Smart cars will automatically sharing real-time data to improve commutes for everyone. Vehicles will use their sensors and cameras to offer details on free parking spaces, traffic conditions, and road hazards.There's so much it can tell you by itself through their car system and services.

But the solution won't give you a truly comprehensive view of the road until there are plenty of connected cars on the road. For sure, currently we are in this transition period, and the transition period can be painful.

Is there a solution can help us smoothly and safely jump into the future?Yes !

First, there is another term need to be introduced. “Vehicle to Infrastructure” which is through the communicative technologies embedded the roadways and other infrastructures and systems, let the road talk to your car constantly.

It is easy to predict in the near future there are both autonomous vehicles and non-autonomous vehicles on the roadways, Get the hardware and system upgraded infrastructures more involved in the ecosystem could be helpful. Those infrastructures installed with camera, sensor, radar connected with public wifi like LinkNYC could be the eyes on the road constantly communicate with the system running in the smart or autonomous vehicles on the road, notice and inform them in advance to avoid the potential risk, improving the roadway safety, efficiency, congestion mitigation and driving experience.

#Thesis Sva-Ixd #solution #pitch

Google’s self-driving car has been involved in its worst crash yet

Google has confirmed that one of its self-driving vehicles sustained significant damage in a traffic accident involving a commercial van on Friday, September 23.

The collision took place at the corner of W El Camino Real and Calderon Ave in Mountain View, California, when the driver of the van ran a red light and struck the passenger side of Google’s Lexus RX 450 SUV.

The car was operating in its autonomous mode at the time of the crash, but that didn’t stop the human driver from slamming on the breaks when he spotted the oncoming van. Unfortunately, his reaction was a little too slow.

The good news, however, is that no one sustained any injuries during the incident. Although, understandably the Google employee in the heavily modified RX 450 was visibly shaken.

It’s worth noting that this isn’t the first time that one of Google’s autonomous vehicles have been caught up in a traffic accident.

Earlier this year, a Google-branded self-driving car was involved in its first and only at-fault accident when it hit a public transit bus while attempting to navigate around a sandbag in the middle of the road.

This latest incident wasn’t the autonomous car’s fault, though. In a statement issued to 9to5Google, the company revealed that the traffic light was green for at least six seconds before the vehicle entered the intersection.

It then went on to add that “thousands of crashes happen everyday on U.S. roads, and red-light running is the leading cause of urban crashes in the U.S. Human error plays a role in 94% of these crashes, which is why we’re developing fully self-driving technology to make our roads safer.”

#Thesis Sva-Ixd #reading research

Trending Blogs

Recently Viewed Blogs

Street Smart - 2017 Ixd Thesis