Tag Archives: autonomous cars

Ericsson Mobility Report Sees a 5G Transport Revolution

logo5g

Remote operation of vehicles with 5G

In the near future it will be a common occurrence to see driverless buses on city streets. A key step towards introducing autonomously driven buses into the public transport system is the development of remote monitoring and control capabilities, which will help to ensure safety, the Ericsson Mobility Report describes.


Key highlights

  • Scania has a 5G proof-of-concept test network used to carry out  trials in which a remote operator drives a bus around its test track
  • 5G will bring a number of benefits to remote control systems, including core network slicing that will enable priority service provisioning, and radio access to bring ultra-low latency and beamforming for high throughput and capacity
  • The work on remotely-controlled buses is making safe, autonomous vehicles a reality

While autonomous vehicles could revolutionize mass transportation as we know it, their safety has been widely debated. To address this concern, remote operation brings a safety mechanism that allows public buses to be monitored and controlled by a remote operator from a distance, if needed. The vision of operators scanning screens and on-hand to intervene if necessary, should contribute to public acceptance of autonomous vehicles.

Network requirements for remote operation include broad coverage, high data throughput and low latency to enable continuous video streaming and to send commands between a remote operations center and a vehicle. 5G will bring a number of benefits to remote control systems, including core network slicing that will enable priority service provisioning, and radio access to bring ultra-low latency and beamforming for high throughput and capacity.

At its headquarters in Södertälje, Sweden, Scania has a 5G proof-of-concept test network devoted to controlling a bus remotely from a vehicle operations center. Work at the site is focused on two important areas: total system response time for remote monitoring and control, and the automated tools required to provision prioritized network services.1

The tests involve a remote operator driving a bus around the test track, as well as to and from the parking facilities. Sensor data from the bus, including a high-resolution video feed, is streamed to the remote operations center over LTE radio access with an evolved 5G core network. The testbed features automated service ordering and provisioning, allowing the set-up and take-down of prioritized network resources needed for the remote monitoring and operation.

The 5G proof-of-concept network is not the biggest source of latency in the complete remote operation system. Additional factors that cause delays include servo-driven mechanics, as well as video encoding and decoding

Isolating and measuring contributors to system response time

A key objective is to isolate and measure the different contributors to the remote control system response time, including network latency. Response time is measured in milliseconds (ms) – for example, from when an operator sees an obstacle on the road and reacts by using the remote controls to apply the brakes, to the point when the result (the bus slowing) is visible in the video presented to the operator. During the tests, total system response times of around 185 ms were achieved. The most significant contributors to the response time and its variation were mechanical delays (physical actuators controlling the bus), followed by the video processing delay and, finally, the network delay (round trip time (RTT)).

Network RTT mostly stayed under 50 ms during the study, although some areas with obstacles along the test route increased the latency beyond this value.  Uplink throughput, which is critical for remote operation, was also measured. In good coverage areas, the uplink throughput was between 10–20 Mbps.

Graph: Illustration of system response time

5G radio access lowers round trip time to under 4 ms

Reducing system response time

Improvements are being made in all areas affecting system response time. Network latency improves significantly with 5G radio access, lowering network RTT to under 4 ms. Video encoding and decoding are on a track to continuously improve, with advances in both codecs and adaptive streaming mechanisms. Mechanical delays will decline as the buses themselves are specially designed for autonomy and remote operation is deployed – rather than being converted from today’s driver-controlled buses.

Automated network resource prioritization

A critical requirement for a remote vehicle control center is the ability to prioritize network services if remote operation is required, through a service ordering API to a mobile telecom operator. In the testbed, an interface based on the Open Mobile Alliance candidate standard is used for network resource prioritization.

Technology is being developed to enable a self-service portal, allowing network customers, such as public transport companies, to specify quality of service (QoS) requirements on their own terms; for example, to prioritize 4K video traffic for 40 buses. The software will then translate this specification into instructions for network resource prioritization.

Mechanical delays will decline as the buses themselves are specially designed for autonomy and remote operation is deployed – rather than being converted from today’s driver-controlled buses

Research concept vehicle

Parallel to the Scania activities, the remote operation of a research concept vehicle (RCV) – developed and custom-built by the Integrated Transport Research Lab at KTH Royal Institute of Technology – was demonstrated at Mobile World Congress 2017. A 5G testbed radio, using a 15 GHz carrier frequency, provided sufficient bandwidth for remotely operating multiple vehicles in the same cell. Delivering the throughput on 15 GHz is accomplished using beamforming; that is, tracking the moving vehicle and focusing the radio power for maximum effect. Due to the low latency of the 5G radio access, RTT was under 4 ms.

Attributes of 5G – including network slicing and low latency – will make safe public transport using autonomous vehicles a reality

The work on remotely-controlled buses and the RCV are making safe, autonomous vehicles a reality. Additionally, the insights from these activities can be applied to other industrial use cases that require high uplink throughput, low network latency and automated service provisioning.

Integrated Transport Research Lab: The Integrated Transport Research Lab (ITRL), a collaboration between the Swedish Royal Institute of Technology (KTH), Scania and Ericsson, is actively exploring intelligent transport systems. Recent studies illustrate the use of emerging 5G cellular technologies in testbed environments.

Graph: Illustration of the Remote control system

Testbed and methodology

Testbed network

The Scania testbed network uses LTE radio access on band 40 (2.3 GHz TDD) to provide data connectivity to the bus. Throughput and RTT were measured between the remotely-operated bus and the network provisioning system. RTT was measured from the remote operations center to the vehicle and back, along with covered radio and network transport (both uplink and downlink). RTT measurements were collected at a rate of one per second, resulting in hundreds of measurements collected from different areas of the test track.

Video link

A single camera at the front of the bus requires an uplink throughput of 8 Mbps to stream a 1080p 60-frames-per-second video from the bus to the operations center. An industrialized solution will include cameras to capture video from the front, back and sides of the bus, requiring about 24 Mbps bandwidth using current codecs.  The video latency was measured using two GPS-synchronized clocks. Each clock displayed time in binary format as a line of LEDs, allowing sub-millisecond resolution. One clock was placed on the bus in view of the camera, while the second clock was attached to the video display at the control center. Photos showing both clocks were taken at one-second intervals. The difference between the clocks indicated the video latency, measured to a high degree of accuracy.

Mechanical controls

The latency of the remote control rig can be reduced to under 1 ms by using optimized equipment. On the other hand, the vehicle latency, including mechanical controls, has a wide variation.

Automated network service prioritization

A cloud-hosted application function (AF) dynamically sets up virtual connections between vehicles and the 5G Evolved Packet Core (EPC) network, with specific QoS attributes such as designated latency levels and guaranteed throughput. This application functionality can be securely opened to third parties through an API. In the use case described here, the testbed uses this API to set up priority virtual connections for vehicles in need of remote operator assistance.

In 5G radio, uplink/downlink allocation will be more flexible and able to meet the demands of uplink critical use-cases

1Ericsson Research, “5G teleoperated vehicles for future public transport” (2017), Rafia Inam, Keven (Qi) Wang, Nicolas Schrammar, Athanasios Karapantelakis, Leonid Mokrushin, Aneta Vulgarakis Feljan, Viktor Berggren and Elena Fersman: www.ericsson.com/research-blog/5g/5g-teleoperated-vehicles-future-public-transport/

Source: Ericsson

Germany, France agree on transnational self-driving test zone

In a bid to win the race to develop self-driving cars and give a leg-up to their automakers over their rivals from Silicon Valley, the two EU nations have joined hands to test the cars on a stretch of road linking them.

Daimler Trucks LKW autonomes Fahren (Daimler AG - Global Communications Commercial Vehicles)

European neighbors Germany and France plan to test self-driving vehicles on a section of road linking the two countries, the transport ministry in Berlin said Wednesday.

The route stretches around 70 kilometers (43 miles), from Merzig in Germany’s western Saarland state to Metz in eastern France. It is aimed at testing “automated and connected driving in real cross-border traffic,” the ministry noted.

“Manufacturers will be able to test the connectivity of their systems, for example when lanes or speed limits change at the border,” German Transport Minister Alexander Dobrindt said in a statement following a meeting with his French counterpart Alain Vidalies.

“We want to set worldwide standards for this key technology through cooperation between Europe’s two biggest car-producing countries,” he added.

The route will allow testing of 5G wireless communications between cars and infrastructure, automated maneuvers such as overtaking and braking, and emergency warning and call systems, among others.

An automated future?

Germany, home to car giants such as Volkswagen, Daimler and BMW, already boasts a number of test zones for automated vehicles on motorways and in cities, but this is the first that will cross into another country.

The transport ministry has offered 100 million euros ($107 million) in funding for the projects.

The tests come as the nation’s traditional carmakers are racing to catch up to Silicon Valley newcomers such as Tesla, Uber and Google parent company Alphabet in the new field, seen as the future of driving.

Automated trucks in particular are expected to shake up the road transport sector in the years to come.

In a glimpse of what lies ahead, manufacturers took part in an experiment last year that saw six convoys of “smart” trucks cross several European countries using “platooning,” in which a leading truck sets the route and speed for wirelessly-connected self-driving followers.

sri/uhe (AFP, Federal Ministry of Transport and Digital Infrastructure)

Source: Deutsche Welle

When will we see driverless cars on UK roads? Lords to investigate

google_self-driving_car_horn_honk

 

The House of Lords Science and Technology Committee will continue its inquiry investigating driverless vehicles on Tuesday 8 November. The Committee will hear evidence from European Officials and industry experts.

This session provides an opportunity for the Committee to hear from representatives from three driverless car trials in Greenwich, Bristol and Milton Keynes. The Committee will be able to explore the progress being made by the trials and the issues they have highlighted relating to the deployment and regulation of driverless cars as well as social and behavioural issues.

The Committee will also examine the extent to which the UK will have to align itself with future international regulation for self-driving vehicles in areas such as cyber-security and data handling and will assess what progress has been made in European and global regulation of autonomous vehicles.

At 10:40am the Committee will hear from:

  • Ms Claire Depré, Head of Sustainable & Intelligent Transport Unit, DG Transport and Mobility
  • Dr Hermann Meyer, CEO, ERTICO –EUROPE
  • Mike Hawes, Chief Executive, Society of Motor Manufacturers & Traders

The Committee are likely to ask:

  • What can European organisations deliver that individual Member States or organisations cannot deliver on their own?
  • What ways is it possible to avoid a situation where European countries have their own individual approach to cybersecurity and privacy requirements for highly autonomous vehicles?
  • To what extent can the UK devise its own regulations and standards?

At 11:40am the Committee will hear from:

  • Professor Nick Reed, Greenwich Automated Transport Environment
  • John McCarthy, Bristol Driverless Cars Project
  • Brian Matthews, Head of Transport Innovation, Milton Keynes Council

The Committee are likely to ask:

  • Are you demonstrating a scientific or engineering process or testing elements of a system to be deployed?
  • Has there been modelling or simulation of deployment on a network of a mixed fleet of non-highly and fully-automated vehicles?
  • Can these new types of vehicle operate safely, efficiently and effectively on current infrastructure or will there have to be significant new infrastructure investment?

The evidence session will take place in Committee room 4A on Tuesday 8 November in the House of Lords at 10:30am.

Source: UK Parliment

UK Autodrive Completes First Collaborative Autonomous Vehicle Trials

Jaguar Land Rover's vehicle completes an automated overtake during Friday's demonstration.

Jaguar Land Rover’s vehicle completes an automated overtake during Friday’s demonstration.

 

The UK’s first collaborative trials of connected and autonomous vehicle technology were successfully completed on Friday, as UK Autodrive partners Jaguar Land Rover, Ford and Tata Motors European Technical Centre (TMETC) jointly demonstrated a number of future vehicle technologies at HORIBA MIRA’s Proving Ground in Nuneaton.

Friday’s demonstration came at the end of a fortnight trials, in which the three vehicle manufacturers were able to successfully demonstrate the programme’s first two connected car features.

The first demonstration showcased cars that can warn their drivers when another connected car up ahead has braked severely, lowering the risk of rear-end collisions when the driver’s view is obscured, for example, by fog or other vehicles.

The second demonstration showed how connected cars can be sent information from traffic lights, allowing them to reduce the likelihood of meeting red lights – potentially improving future traffic flow and lowering emissions in urban areas.

“There has already been a lot of public focus on self-driving vehicles, but connected car technology may be just as revolutionary,” said Tim Armitage, Arup’s UK Autodrive project director.

“The benefits of having cars that can communicate with each other and their surroundings could be very significant – from increased road safety to improved traffic flow, more efficient parking and better information for drivers.”

Jaguar Land Rover also used Friday’s event to demonstrate a self-driving Range Rover Sport that was able to overtake slower moving vehicles automatically – and also reject overtake requests if it detects another vehicle in the occupant’s “blind spot”.

A Ford test driver receives a brake light warning triggered by the Jaguar ahead of him.

A Ford test driver receives a brake light warning triggered by the Jaguar ahead of him.

The UK Autodrive demonstration was also welcomed the UK’s Roads Innovation Minister, John Hayes, who said: “This technology has the potential to revolutionise travel by making journeys safer and cutting congestion for motorists. I’m proud that the UK is a world leader when it comes to developing connected and automated vehicles, and we are further establishing ourselves as the place to test and invest in this emerging technology.”

 

Further UK Autodrive trials and demonstrations are scheduled to take place at HORIBA MIRA during the spring of next year, before moving out onto closed-off areas of Milton Keynes and Coventry in late 2017. The project will culminate in a series of open road trials and demonstrations to be held in both cities in 2018.

Jaguar Land Rover's vehicle completes an automated overtake during Friday's demonstration.

Jaguar Land Rover’s vehicle completes an automated overtake during Friday’s demonstration.

The driver of the TMETC vehicle receives a speed advisory to help reach a green light.

The driver of the TMETC vehicle receives a speed advisory to help reach a green light.

JAGUAR LAND ROVER DRIVES FORWARD CONNECTED AND AUTONOMOUS VEHICLE TECHNOLOGIES

 

crophero-advancedhighwayassist2-1200x462

Jaguar Land Rover showcased its latest Connected and Autonomous Vehicle technologies as part of the UK Autodrive demonstrations taking place at HORIBA MIRA today.

 

In a UK first Jaguar Land Rover collaborates with other manufacturers to trial new connected technologies that allow cars to talk to each other

  • Connected and Autonomous Vehicle technologies will help make driving safer and cleaner in the future
  • Showcasing three research technologies, including a car that knows what speed to travel at to ensure traffic lights are always on green

Coventry, UK: Jaguar Land Rover showcased its latest Connected and Autonomous Vehicle technologies as part of the UK Autodrive demonstrations taking place at HORIBA MIRA today. In a UK first, Jaguar Land Rover is working with Ford and Tata Motors European Technical Centre to test connected technologies that will allow cars to talk to each other as well as the roadside infrastructure, such as traffic lights, in the future.

Connected and Autonomous Vehicle technologies are one of Jaguar Land Rover’s research priorities. It is creating a fleet of more than 100 research vehicles to develop and test a wide range of Connected and Autonomous Vehicle technologies over the next four years. Ultimately, these technologies will enhance the driving experience as well as making driving smarter, safer and even cleaner in the years to come.

We know that there’s a huge potential for these technologies in future vehicles around the world. Until now we have focused on communication between Jaguar and Land Rover vehicles, this collaborative approach is a major stepping stone towards all Connected and Autonomous Vehicles co-operating with each other in the future. Our aim is to give drivers exactly the right information at the right time and collaborations with other manufacturers are essential to help us deliver this commitment to our customers.

TONY HARPER
HEAD OF RESEARCH, JAGUAR LAND ROVER,

Jaguar Land Rover is developing both fully and semi-autonomous vehicle technologies to help the driver with the challenging or more tedious parts of driving whilst maintaining an enjoyable driving experience. The company’s vision is to make the autonomous car viable in the widest range of real life, on and off road driving environments and weather conditions.

Connected and Autonomous Vehicle Technologies

With Advanced Highway Assist the vehicle can overtake vehicles automatically as well as stay in its lane on the motorway without the driver having to touch either the steering wheel or the pedals.

Electronic Emergency Brake Light Assist warns drivers when a vehicle ahead brakes severely or unexpectedly. This is particularly useful when driving in dense fog or if the vehicle in front is out of sight.

Imagine travelling across central London or Paris without needing to stop at traffic lights because they are always on green. This could be possible with Green Light Optimal Speed Advisory. The car connects to traffic lights advising the driver of the best driving speed required to reach the lights when they are on green. This will improve traffic flow, CO₂ emissions as well as the driver’s experience.

UK Autodrive

UK Autodrive is a consortium of leading technology and automotive businesses, local authorities and academic institutions working together on a three year UK trial of self-driving vehicle and connected car technologies. It is helping to establish the UK as a global hub for the research, development and integration of automated and connected vehicles into society. It will also investigate other aspects of automated driving, including safety and cyber-security issues as well as the public’s acceptance for connected and autonomous vehicles.

Source: JLR

 

 

 

Autonomous cars expected to hit 24 million units by 2030, says Berg Insight

(c)iStock.com/chombosan

A new research report from Berg Insight says the first autonomous cars will debut in 2020, and predicts that the total number of new registrations of autonomous cars will grow at a CAGR of 62% from 0.2 million units in 2020 to reach 24 million units in 2030.

As the analyst firm explains, cars are among the costliest as well as one of the most inefficiently used assets today, but when operated around the clock on a service based business model can results in a tremendous potential increase of their utilisation rate. Moreover, autonomous cars will potentially improve the quality of life for the people who are unable to drive, reduce the number of fatalities and accidents in road traffic and increase overall traffic efficiency.

That’s the theory, at any rate. The research notes that the active installed base of autonomous cars is estimated to have reached about 71 million at the end of 2030, including SAE Level 3 and 4 cars.

The report states that the key to the development of self-driving cars lies in the software that will interpret sensor information and manage the driving logic. In order to further develop the autonomous technology, many new actors such as IT firms and other technology-oriented companies have joined incumbent automakers who have already initiated projects to develop self-driving features in their cars.

Ludvig Barrehag, M2M/IoT analyst at Berg Insight, said: “These pathways do not contradict each other as different autonomous systems are suitable in different use cases. We will continue to see development from both sides for still some years before the two approaches converge.”

Source: Connected Car/Berg Insight

nuTonomy launches World’s first public trial of autonomous Taxi’s in Singapore

 

press_car

nuTonomy Launches World’s First Public Trial of
Self-Driving Car Service and Ride-Hailing App

nuTonomy, the leading developer of state-of-the art software for self-driving cars, today launched the first-ever public trial of a robo-taxi service. The trial, which will continue on an on-going basis, is being held within Singapore’s one-north business district, where nuTonomy has been conducting daily autonomous vehicle (AV) testing since April.

Beginning today, select Singapore residents will be invited to use nuTonomy’s ride-hailing smartphone app to book a no-cost ride in a nuTonomy self-driving car that employs the company’s sophisticated software, which has been integrated with high-performance sensing and computing components. The rides will be provided in a Renault Zoe or Mitsubishi i-MiEV electric vehicle that nuTonomy has specially configured for autonomous driving. An engineer from nuTonomy will ride in the vehicle to observe system performance and assume control if needed to ensure passenger comfort and safety.

Throughout the trial, nuTonomy will collect and evaluate valuable data related to software system performance, vehicle routing efficiency, the vehicle booking process, and the overall passenger experience. This data will enable nuTonomy to refine its software in preparation for the launch of a widely-available commercial robo-taxi service in Singapore in 2018.

Earlier this month, nuTonomy was selected by the Singapore Land Transport Authority (LTA) as an R&D partner, to support the development of a commercial AV service in Singapore. This trial represents the first, rapid result of this partnership. nuTonomy is the first, and to date only, private enterprise approved by the Singapore government to test AVs on public roads.

CEO and co-founder of nuTonomy, Karl Iagnemma, said, “nuTonomy’s first-in-the-world public trial is a direct reflection of the level of maturity that we have achieved with our AV software system. The trial represents an extraordinary opportunity to collect feedback from riders in a real-world setting, and this feedback will give nuTonomy a unique advantage as we work toward deployment of a self-driving vehicle fleet in 2018.”

In May of this year, nuTonomy completed a $16M Series A funding led by Highland Capital Partners that included participation from Fontinalis Partners, Signal Ventures, Samsung Ventures, and EDBI, the dedicated corporate investment arm of the Singapore Economic Development Board.

In addition to Singapore, nuTonomy is operating self-driving cars in Michigan and the United Kingdom, where it tests software in partnership with major automotive manufacturers such as Jaguar Land Rover.

To see nuTonomy in action please click the link below:

https://youtu.be/Fp13jIr8Y7E

Source: nuTonomy

 

 

Volvo Cars to launch UK’s largest and most ambitious autonomous driving trial

volvo autonomous Volvo Cars, the premium car maker, is to begin the UK’s most ambitious autonomous driving trial next year to speed up the introduction of a technology that promises to massively reduce car accidents as well as free up congested roads and save drivers valuable time.

 The Swedish company, whose name is synonymous with automotive safety ever since it invented the three point seat belt in 1959, is pioneering the development of autonomous driving systems globally as part of its commitment that no one will be seriously injured or killed in a new Volvo by the year 2020.

 “Autonomous driving represents a leap forward in car safety,” said Håkan Samuelsson, president and chief executive. “The sooner AD cars are on the roads, the sooner lives will start being saved.”

Mr Samuelsson will make his comments at a seminar sponsored by Volvo and Thatcham, the insurance industry’s research organisation, in London on May 3 entitled ‘A Future with Autonomous Driving Cars – Implications for the Insurance Industry’ at the America Conference Centre in London.

 Volvo’s UK-based test will be called ‘Drive Me London’ and will differentiate itself from other AD programmes by using real families driving AD cars on public roads.

Volvo will source its data from these everyday users and use this data to develop AD cars that are suitable for real world driving conditions, rather than the more unrealistic conditions found on test tracks. Thatcham Research will be providing the technical data analysis and any professional test drivers needed as part of the trial.

 Drive Me London will begin in early 2017 with a limited number of semi-autonomous driving cars and expand in 2018 to include up to 100 AD cars, making it the largest and most extensive AD testing programme on Britain’s streets.

 The introduction of AD cars promises to revolutionise Britain’s roads in four main areas – safety, congestion, pollution and time saving.

 Independent research has revealed that AD has the potential to reduce the number of car accidents very significantly, in some cases by up to 30 per cent. Up to 90 per cent of all accidents are presently caused by driver error or distraction, something that should largely disappear with AD cars.

 “Vehicle manufacturers are predicting that highly autonomous vehicles, capable of allowing the driver to drop ‘out of the loop’ for certain sections of their journey, will be available from around 2021. Without doubt, crash frequency will also dramatically reduce. We’ve already seen this with the adoption of Autonomous Emergency Braking (AEB) on many new cars. Research in the US by NHTSA predicts that by 2035, as a result of autonomous and connected cars, crashes will be reduced by 80%. Additionally, if a crash unfortunately can’t be avoided, then the impact speed will also drop as a result of the system’s performance – reducing the severity of the crash,” said Peter Shaw, chief executive at Thatcham Research.

 “Driverless cars will see our journeys become faster, cleaner and safer. The UK is leading the way in developing the technology needed to make this a reality thanks to our world-class research base and these types of trials will become increasingly common,” Sajid Javid, UK Secretary of State for Business, Innovation and Skills said. “Such advances in technology prove the fourth industrial revolution is just around the corner and our determination to be at the forefront is why we are attracting top names from across the globe for real-world testing.”

 In terms of congestion, AD cars allow traffic to move more smoothly, reducing traffic jams and by extension cutting dangerous emissions and associated pollution. Lastly, reduced congestion saves drivers valuable time.

 “There are multiple benefits to AD cars,” said Mr Samuelsson. “That is why governments globally need to put in place the legislation and infrastructure to allow AD cars onto the streets as soon as possible. The car industry cannot do it all by itself. We need governmental help.”

Source: Volvo

Toyota Opens Third Autonomous Driving Research Facility

TOYOTA MOVES FURTHER DOWN THE ROAD TO Toyota-LogoAUTONOMOUS DRIVING WITH ANNOUNCEMENT OF THIRD RESEARCH FACILITY

Toyota Research Institute to fund studies at the University of Michigan

Toyota is to open a third Toyota Research Institute facility in the USA, which will fund studies into artificial intelligence, materials science and robotics. It will be located in Ann Arbor, close to the University of Michigan campus.

The new centre, known as TRI-ANN, is due to open in June this year and is set to build a 50-strong team. It will join the TRI facility which opened in Palo Alto in January (TRI-PAL) to work with Stanford University, and TRI-CAM in Cambridge, which works with the Massachusetts Institute of Technology.

Dr Gill Pratt, TRI Chief Executive, speaking at the GPU Technology Conference in San Jose yesterday (7 April) explained the decision to locate at Ann Arbor reflected the links Toyota has with the community and the benefits to be gained from close proximity to the university and facilities such as the Toyota-sponsored Mobility Transformation Center.

Toyota also has two well-established technical centres nearby, which have been researching the concept of autonomous vehicles for more than a decade. A group of about 15 team members will transfer to the new TRI-ANN facility when it opens, together with University of Michigan professors Ryan Eustice and Edwin Olson, who will be area leads respectively for mapping/localisation and perception.

Professor Olson commented: “Sensor hardware and algorithms are improving at a tremendous pace. TRI researchers will push the frontier even further, resulting in safer vehicles and more helpful robots in the home.”

Professor Eustice added: “Ann Arbor is a fantastic location for TRI to expand its autonomous driving efforts. We will benefit from Toyota’s existing team and the University of Michigan’s research talent and facilities where we can perform extreme-limit testing in a wide variety of environments.”

Each TRI facility will have a different core discipline: TRI-ANN will focus primarily on fully autonomous (chauffeured) driving; TRI-PAL will work on so-called “guardian angel” driving, where the driver is always engaged, with the vehicle assisting when needed; and TRI-CAM dedicating a large part of its work to simulation and deep learning.

The Toyota Research Institute is an enterprise designed to bridge the gap between fundamental research and product development.  With initial funding of $1 billion, it has four initial mandates.

The first is to strive to enhance automobile safety with the ultimate goal of creating a car that is incapable of causing a crash, regardless of the skill or condition of the driver.

The second is to work to increase access to vehicles for those who otherwise could not drive, including older people and those with special needs.

The third is to help translate Toyota’s expertise in creating products for outdoor mobility into products for indoor mobility.

Finally, TRI will accelerate scientific discovery by applying techniques from artificial intelligence and machine leaning, particularly in the sphere of materials science. This will also help reduce costs and improve the performance of future mobility systems.

Beyond these, TRI is engaging in multiple projects with the three universities and is also pursuing collaboration with other car makers, IT companies, suppliers, research labs and academic institutions for the joint development of autonomous technologies.

Dr Pratt said: “Although the industry, including Toyota, has made great strides in the last five years, much of what we have collectively accomplished has been easy, because most driving is easy. Where we need autonomy to help most is when the driving is difficult.  It’s this hard part that TRI intends to address.

“Toyota’s goal is safe mobility for all, at any time, in any place, and the tremendous improvements in quality of life that such universal mobility can bring.”

Source: Toyota

Ford Conducts Autonomous Snow Driving Tests

FORD CONDUCTS INDUSTRY-FIRST SNOW TESTS OF AUTONOMOUS VEHICLES — FURTHER ACCELERATING DEVELOPMENT PROGRAM

ford-conducts-industry-first-snow-tests-of-autonomous-vehicles.img.1452519750506

  • Ford is the first automaker to test fully autonomous vehicles in winter weather, including snow – a major step toward fully autonomous driving.
  • Ford’s fully autonomous vehicle strategy uses high-resolution 3D mapping and LiDAR for localization to facilitate driving when road markings are not visible
  • Winter weather testing strengthens Ford leadership in autonomous vehicle development, building on recent news of the company’s expansion of its fully autonomous vehicle fleet – now the largest of all automakers

 

Ford is conducting the industry’s first autonomous vehicle tests in snow-covered environments – a major step in the company’s plan to bring fully autonomous vehicles to millions of customers worldwide.

Unlike other major car manufacturers and technology companies, which have tested autonomous vehicle technology only in dry, mostly sunny climates, Ford knows the future of autonomous driving cannot rely on ideal conditions.

“It’s one thing for a car to drive itself in perfect weather,” said Jim McBride, Ford technical leader for autonomous vehicles. “It’s quite another to do so when the car’s sensors can’t see the road because it’s covered in snow. Weather isn’t perfect, and that’s why we’re testing autonomous vehicles in wintry conditions – for the roughly 70 percent of U.S. residents who live in snowy regions.”

Ford’s winter weather testing takes place in Michigan, including at Mcity – a 32-acre, full-scale simulated real-world urban environment at the University of Michigan.

Fully autonomous driving can’t rely on GPS, which is accurate only to several yards – not enough to localize or identify the position of the vehicle. And it’s key that an autonomous vehicle knows its precise location, not just within a city or on a road, but in its actual driving lane – a variation of a few inches makes a big difference.

LiDAR, on the other hand, is much more accurate than GPS – identifying the Fusion Hybrid’s lane location right down to the centimeter. LiDAR emits short pulses of laser light to precisely allow the vehicle to create a real-time, high-definition 3D image of what’s around it.

In ideal weather, LiDAR is the most efficient means of gathering important information and metadata – underlying information about the data itself – from the surrounding environment, sensing nearby objects and using cues to determine the best driving path. But on snow-covered roads or in high-density traffic, LiDAR and other sensors such as cameras can’t see the road. This is also the case when the sensor lens is covered by snow, grime or debris.

Undaunted by this challenge, Ford and University of Michigan technologists began collaborating toward a solution that would allow an autonomous vehicle to see on a snow-covered road.

How snow autonomy works
To navigate snowy roads, Ford autonomous vehicles are equipped with high-resolution 3D maps – complete with information about the road and what’s above it, including road markings, signs, geography, landmarks and topography.

“Maps developed by other companies don’t always work in snow-covered landscapes,” said Ryan Eustice, associate professor at University of Michigan college of engineering. “The maps we created with Ford contain useful information about the 3D environment around the car, allowing the vehicle to localize even with a blanket of snow covering the ground.”

An autonomous vehicle creates the maps while driving the test environment in favorable weather, with technologies automatically annotating features like traffic signs, trees and buildings. When the vehicle can’t see the ground, it detects above-ground landmarks to pinpoint itself on the map, and then subsequently uses the map to drive successfully in inclement conditions.

“The vehicle’s normal safety systems, like electronic stability control and traction control, which often are used on slippery winter roads, work in unison with the autonomous driving software,” said McBride. “We eventually want our autonomous vehicles to detect deteriorating conditions, decide whether it’s safe to keep driving, and if so, for how long.”

A pioneer in autonomy
Winter driving still presents a host of challenges, but Ford’s testing marks an important achievement on the road to autonomous driving. That road goes back roughly a decade, to the first-generation autonomous vehicle from Ford – a LiDAR-equipped F-250 Super Duty.

In 2013, Ford launched its second-generation autonomous vehicle platform, a Fusion Hybrid sedan using more advanced LiDAR sensors. This past summer, Ford transitioned its fully autonomous vehicle development program from the research to advanced engineering phase, the second of three phases before entering production.

Earlier this month, Ford announced it is taking the next step – tripling its fully autonomous development fleet to 30 vehicles being tested on roads and test tracks in California, Arizona and Michigan. This makes the company’s fully autonomous vehicle fleet the largest of all automakers. These third-generation autonomous vehicles continue to be based on a Fusion Hybrid sedan now featuring the first auto-specific LiDAR sensor capable of handling different driving scenarios – thanks in part to its longer range of around 200 meters.

Building on more than a decade of Ford autonomous vehicle research, this advancement is a key element of Ford Smart Mobility – the plan to take Ford to the next level in connectivity, mobility, autonomous vehicles, the customer experience, and data and analytics.

Source: Ford