# AI News, Driverless cars change lanes more like humans do

## Driverless cars change lanes more like humans do

At the International Conference on Robotics and Automation tomorrow, researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) will present a new lane-change algorithm that splits the difference.

It allows for more aggressive lane changes than the simple models do but relies only on immediate information about other vehicles' directions and velocities to make decisions.

'The optimization solution will ensure navigation with lane changes that can model an entire range of driving styles, from conservative to aggressive, with safety guarantees,' says Rus, who is the director of CSAIL.

The buffer zones describe not only the vehicles' current positions but their likely future positions within some time frame.

For any given method of computing buffer zones, algorithm designers must prove that it guarantees collision avoidance, within the context of the mathematical model used to describe traffic patterns.

With the MIT researchers' system, if the default buffer zones are leading to performance that's far worse than a human driver's, the system will compute new buffer zones on the fly -- complete with proof of collision avoidance.

Using the static, precomputed buffer zones would only allow for conservative driving, whereas our dynamic algorithm allows for a broader range of driving styles.'

## Making driverless cars change lanes more like human drivers do

In the field of self-driving cars, algorithms for controlling lane changes are an important topic of study.

But most existing lane-change algorithms have one of two drawbacks: Either they rely on detailed statistical models of the driving environment, which are difficult to assemble and too complex to analyze on the fly;

“The optimization solution will ensure navigation with lane changes that can model an entire range of driving styles, from conservative to aggressive, with safety guarantees,” says Rus, who is the director of CSAIL.

For any given method of computing buffer zones, algorithm designers must prove that it guarantees collision avoidance, within the context of the mathematical model used to describe traffic patterns.

With the MIT researchers’ system, if the default buffer zones are leading to performance that’s far worse than a human driver’s, the system will compute new buffer zones on the fly — complete with proof of collision avoidance.

“The autonomous vehicles were not in direct communication but ran the proposed algorithm in parallel without conflict or collisions,” explains Pierson. “Each car used a different risk threshold that produced a different driving style, allowing us to create conservative and aggressive drivers.

Using the static, precomputed buffer zones would only allow for conservative driving, whereas our dynamic algorithm allows for a broader range of driving styles.” This project was supported, in part, by the Toyota Research Institute and the Office of Naval Research.

## Making Driverless Cars Change Lanes More Like Human Drivers Do

A new algorithm developed by the Massachusetts Institute of Technology&#39;s (MIT) Computer Science and Artificial Intelligence Laboratory (CSAIL) could improve the way driverless vehicles change lanes.

If default buffer zones are leading to performance that is significantly worse than a human driver&#39;s, the system will compute new buffer zones on the fly, with proof of collision avoidance.

Based on estimates of the car&#39;s direction and velocity, the system generates a logistic function that, when multiplied by the Gaussian distribution, skews the distribution in the direction of the car&#39;s movement.

## New Algorithm Improves the Ability of Autonomous Cars to Change Lanes in Traffic

Current algorithms that help autonomous cars change lanes rely on statistical models of the driving environment around the car.

The algorithms can also be too simple and force the car to make impractical and often apprehensive decisions, which can lead to the car never changing lanes.

The autonomous cars using this new system maintain collision avoidance even though the buffer zones are created in real time.

The new system creates a new logistic function in real time based on the estimated direction and velocity.

The skewed distribution is the key to creating a new buffer zone developed while the car is driving in real time, and the mathematical equation allows the autonomous car to create a buffer zone in real time.

'The autonomous vehicles were not in direct communication but ran the proposed algorithm in parallel without conflict or collisions,' said Alyssa Pierson, a postdoc at CSAIL and first author on the new paper.

'Each car used a different risk threshold that produced a different driving style, allowing us to create conservative and aggressive drivers.

Using the static, precomputed buffer zones would only allow for conservative driving, whereas our dynamic algorithm allows for a broader range of driving styles.'

## Lane-Changing Algorithm Improves Driverless Vehicle Performance

-moz-border-radius:50%}.eg-herbert-hoover-wrapper .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

-webkit-filter:grayscale(100%)}.eg-herbert-hoover-wrapper:hover .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='1 0 0 0 0,0 1 0 0 0,0 0 1 0 0,0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

-webkit-filter:grayscale(0%)}.eg-lyndon-johnson-wrapper .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

-webkit-filter:grayscale(100%)}.eg-lyndon-johnson-wrapper:hover .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='1 0 0 0 0,0 1 0 0 0,0 0 1 0 0,0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

-webkit-filter:grayscale(100%)}.eg-wilbert-wrapper:hover .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='1 0 0 0 0,0 1 0 0 0,0 0 1 0 0,0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

position:absolute}.eg-howardtaft-wrapper .esg-media-poster,.eg-howardtaft-wrapper .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='1 0 0 0 0,0 1 0 0 0,0 0 1 0 0,0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

-webkit-filter:grayscale(0%)}.eg-howardtaft-wrapper:hover .esg-media-poster,.eg-howardtaft-wrapper:hover .esg-media-poster{filter:url('data:image/svg+xml;utf8,<svg xmlns='http://www.w3.org/2000/svg'><filter id='grayscale'><fecolormatrix type='matrix' values='0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0.3333 0.3333 0.3333 0 0 0 0 0 1 0'></fecolormatrix></filter></svg>#grayscale');

Japex to buy into west coast LNG development

Miners struggling with higher costs, lower prices

FTG Aerospace to supply avionics to Rockwell Collins

Canada keeping up pressure on US for Keystone XL approval

Manufacturing the sole industry showing job losses in February

DART Aerospace re-branding itself to reach wider markets

Bombardier nearly ready to flight test CSeries

Skilled labour shortage in world oil industry: report

GM investing \$250 million at Ingersoll plant

Volkswagen to produce super-efficient hybrid

Bombardier holds update on CSeries aircraft

Russian leasing company orders 42 CSeries jets from Bombardier

RV industry has growing role in Canada's economy: study

Pratt & Whitney Canada announces helicopter engine contracts

BC refinery close to financing deal

SPPCA's new landing gear facility opening in Mississauga

Clean energy expected to surge as pv costs drop

Economy managed slight growth in Q4, but shrank in December

## Google: Human drivers are the problem

Google revealed on Monday that its autonomous vehicles have been in 11 'minor' accidents in California since it launched its self-driving car project over six years ago.

Urmson said in a blog post that Google's cars have been side-swiped 'a couple of times,' rear-ended seven times -- 'mainly at traffic lights but also on the freeway' -- and hit by a car that drove through a stop sign.

'The most common accidents our cars are likely to experience in typical day to day street driving — light damage, no injuries — aren't well understood because they're not reported to police,' Urmson explained.

(An AP report said that two of the recent four accidents happened while the car was in 'self-driving' mode, while the other two incidents took place when a person was in control of the vehicle.)

Google has already identified many patterns of driver behavior, such as lane-drifting and red-light running, that are leading indicators of significant collisions, and programmed its cars to adapt.

For example, Google's cars pause after a light turns green before moving through an intersection to avoid potential red-light runners.

If a driver tries to abruptly cut lanes, Google's cars slow down when they sense another vehicle entering a pre-determined buffer zone to avoid collision.

The company's self-driving cars have driven nearly a million miles autonomously, on both freeways and city streets, and Google says it will continue to test drive its cars to gather more data about common driving problems.

Arduino Garden Controller - Automatic Watering and Data Logging

Gardening in the modern age means making things more complicated and arduous, with electrons, bits, and bytes. Behold: the garduino. My brother got me an ...

Resilience and Security in Cyber-Physical Systems: Self-Driving Cars and Smart Devices

The future will be defined by autonomous computer systems that are tightly integrated with the environment, also known as Cyber-Physical systems (CPS).

2018 Spring Seminar #8: Estimating Drivers’ Behaviors from Event Data Recorder Data

Scott Berkun: "The Year Without Pants" | Talks at Google

A behind-the-scenes look at the firm behind WordPress.com and the unique work culture that contributes to its phenomenal success. Automattic Inc, runs ...

NEW 2018 Audi A8 Development

NEW 2018 Audi A8 Development. The experts at Audi long ago abandoned the obsession with using a single material in lightweight design. With a mix of ...

Energy: Tesla Motors Just Put its Growth Pedal to the Metal *** INDUSTRY FOCUS ***

If Tesla Motors plans for growth and its new Model 3 weren't ambitious enough, it now wants to make it all happen by 2018. Plus, earnings from Royal Dutch ...

Building Science with CMake