Survey  on AI self-driving cars around the world
What’s the future of personal transportation? Well, you’ll likely be spending a lot less time behind the wheel, for one. The rise of self-driving cars means that some scenes out of science-fiction flicks (movie: Total Recall or I, Robot) are now reality—and even more will be available soon.
What is self-drive car?
Self-driving vehicles are cars or trucks in which human drivers are never required to take control to safely operate the vehicle. Also known as autonomous or “driverless” cars, they combine sensors and software to control, navigate, and drive the vehicle.
An autonomous car is a vehicle that is capable of sensing its environment and navigating without human input. Autonomous cars combine a variety of techniques to perceive their surroundings, including radar, laser light, GPS, odometry, and computer vision.
In August 1961, Popular Science reported on the Aeromobile 35B, an air-cushion vehicle (ACV) that was invented by William Bertelsen and was envisioned to revolutionize the transportation system, with personal self-driving hovering cars that could speed up to 1,500MPH.
Let’s, get a look some existing Artificial intelligence self-driving car around the world and how its work?
First of all, get the knowledge of 5 level of self-driven vehicle autonomy given in below table 
level


Name
Narrative Definition
Execution of steering and acceleration/deceleration
Monitoring of driving environment
Fallback performance
Of dynamic driving task
System capability
(driving modes)
0


No automation
The fulltime performance by the human driver of all aspects of the dynamic task, even when enhanced by warning or intervention system
Human driver
Human driver
Human driver
N/A
1


Driver assistance
Driving mode specific execution by driver assistance system of either steering or accele
human driver
human driver
human driver
Some driving
modes
2


Partial Automation
the driving mode specific execution by one or nore driver assistance systems of both steering and accelaration/deceleration using information about the driving environment and with the exepectation thast the human driver perfoem all remeaining aspects of the dynamic driving task.
system
Human driver
Human driver
Some driving modes
*


Automated driving system
(Monitors the driving envinment)








3


Conditional Automation
The driving mode-specific performance by an automated driving system of all aspects of dynamic driving task, even if a human driver does not respond appropiately to request to intervene.
System
System
Human driver
Some driving mode
4


high Automation
The driving mode specific performance by an automated driving system of all aspects of the dynamic driving task, even if a human driver not respond appropriately to a request to intervene
System
system
System
Some driving modes
5


Full Automation
the full-time performance by an automated driving system of all aspects of the dynamic driving task under all roadway and environment condition that can be managed by alumna driver
System
System
System
All driving modes

                                            Table: 1 Levels of vehicle autonomy
1)      GM- General Motors:
An American automobile company GM recently launches level 4 self-driving car “CRUISE AV”. Cruise AV is properly driverless car. it has no steering wheel, no pedals and no real driver at all. Car will hit the road in 2019.
         Intended just for a GM-operated ride-sharing service, you'll summon a Cruise AV using an app, much like using a service like Lyft -- which GM also happens to own a fair bit of. You can input your destination through the app and the car will safely, silently usher you along.
         You can also input preferred temperature, music and passengers will receive constant updates on the car's progress and status through a series of touch screens: one in the dash up front two more behind the headrests.


Figure: 1 sensor system of cruise AV
2)      Way MO:
Waymo is Autonomous company of parent company Google which working on self-driving minivan since 2009. Way mo. was 1st autonomous car company to issue public safety report to federal government outlining their technology’s feature.
         The 3rd level Waymo mini vans have always safety drivers as backups whether or not the vehicle was part of its public pilot.
         They will launch level 4 autonomy Chrysler Pacifica minivan.
         Waymo minivan technology includes 3 LIDAR systems at roof top, one forward facing camera at roof top, radar sensor to detect object in rain, fog or snow at back spoiler, self-driving sensors on front hood and wheel hub sensor at fender.              
Figure: 2 way Mo Chrysler Pacifica
3)      Daimler Bosch:
The parent company Mercedes-Benz partnered with supplier Bosch to crate Level 4 and Level 5 autonomous car by 2020.Daimler already given us a hint of what it believes self-driving cars will be like.
      The F015 concept: Mercedes Benz F015 is only able to follow a pre-programmed route and simple commands on close course.
   Prototype of F015 experienced a technical difficulty half way and needed to be manually driven back to stating bay.
         I-phone running Mercedes-Benz App allow us to remotely call F015 from its nearly parking spot to pick us up and park itself.
         When someone passed in front of car, LEDs in grill glowed to indicate that car had detected person as pedestrian. “GO AHEAD PLEASE”, it said through exterior speakers. Entire dashboard was taken up by ultra-wide gesture-controlled display.
         For interaction between vehicles and environment F015 use extended sensor system to monitor surrounding in 360-degree radius. After system interpret and analyse the data.
         Mercedes- Benz also demonstrated how laser in grill could even project a virtual cross walk onto road to give pedestrians a visual cue and how passengers could wave hand to stopped ROBO car.
                         
                                                       Figure: 3 Mercedes-Benz F015
4)      FORD:
By year 2021 FORD will mass   produce vehicles with full autonomy- no steering wheels, no gas pedal and no driver.
           To develop its autonomous driving capability, Ford will invest in several companies:
   Teledyne: The world's LiDAR leader. Sensors that help vehicles self-navigate
  SAIPS: A machine learning company based in Israel, which uses object detection to help vehicles learn about surroundings.
    Nirenbrg Neuroscience: A device for returning sight to blind patients by cracking neural code that brings sight to the brain.
    Civil Maps: Will use AI to develop 3D, high-resolution maps of surroundings.
    Ford hasn't been as quick to discuss its plans toward developing an autonomous vehicle. Unlike Tesla's, it doesn't have something like Autopilot currently available to drivers. 
         Ford’s hybrid model based on unique approach to use sensor that detect and compare it, to map it.
         In direct approach use the same data but directly interpret object in environment and require more advanced software.
Figure: 4 Ford’s hybrid driverless car
5) Volkswagen Group:
In recent months, Volkswagen partnered with Aurora Innovation to integrate Aurora's self-driving system of sensors, hardware, and software into Volkswagen's vehicles.
         Aurora works at the intersection of rigorous engineering and applied machine learning to address one of the most challenging, important and interesting opportunities of our generation: transforming the way people and goods move.
         Volkswagen’s latest vision of the future is a “subtly wedged shaped” electric and autonomous concept vehicle.
         The fully autonomous concept is called Sadric-a combination of the words self-driving car and has no pedals or steering wheel.
         The vehicle can be summoned with the push of a button and shuttle individuals to their destination, just like the human-operated services Uber and Lyft do today.
         This is the first concept from Volkswagen Group, the parent company of brands that include Volkswagen Passenger Cars, Audi, Bentley, Skoda, and Porsche.
         Sadri signifies the importance of fully automated driving to the automaker, whether it’s in a wedged-shaped shuttle or a feature within a sports car that allows for human or robot control.  
      The control button in Sadri hints at that interest. The button that summons the Sadri is linked to a universal mobility ID. This allows customers to switch between their own personal autonomous Sadri—under one of VW’s brands—to a shared vehicle in another city. 
Figure:5 volks wagon’s autonomous Sadri and its control button
6) BMW-Intel-FCA:
BMW teamed with Intel and camera and sensor company Mobileye in 2016 with the goal of bringing self-driving cars to market by 2021. The three companies are working on the BMW in EXT model
         They plan to have Level 3 cars for consumers in 2021, and Level 4 and 5 ride-hail cars for urban pilot programs in the same year as well.
         BMW is currently testing self-driving cars on public roads, using the brand Personal Co-pilot for the technology.
         BMW's cars will use radar, lidar and cameras, with some technology supplied by its core partners Intel and Mobileye. 
         The cars' computers store high-definition maps of specific roads created by Here, a digital mapping consortium owned by BMW, Audi and Mercedes-Benz. This type of maps not only show the roads, but also permanent objects, such as buildings and traffic infrastructure.
         The car compares what its sensors detect to its stored map, letting it determine its precise location. The map also shows where it is allowed to drive.
Figure: 6 BMW in EXT concept car
7) Aptiv:
Aptiv is a global technology company that develops safer, greener and more connected solutions enabling the future of mobility. 
         in November, Aptiv -formerly Delphi Automotive -purchased autonomy, a leading developer of autonomous driving software. 
         Aptiv recently paired with Lyft at CES 2018 to offer Level 4 autonomous rides down the Las Vegas strip to attendees.
         “Aptiva’s automated driving technology is seamlessly integrated into any design without cumbersome sensors on the top or sides of the vehicle
         LIDAR pucks are visible on or around the bumper fasciae on the front and rear of the vehicle — two other areas commonly damaged during collisions.

Figure:7 Aptiv self-driving system

Figure: 8 Active self-driving concepts in BMW 540i
8) Renault-Nissan-Mitsubishi Alliance:
This alliance is working on an autonomous ride-hailing and car-sharing service, to launch within the next 10 years. It will begin piloting a self-driving taxi service in Japan this year.
         The vehicles will be geofenced along a 4.5-kilometer route between Nissan’s global headquarters and the Yokohama World Porters shopping centre. 
         Safety drivers will remain behind the wheel of the vehicles, and a remote operations centre has been set up to monitor the car trips.
         An in-car tablet screen will show selections of nearly 500 recommended places of interest and events in the vicinity.
         Additionally, about 40 discount coupons for retailers and restaurants in the area are available for download on the participants’ own smart phones.

         Participants will be asked to complete a survey about their overall user experience, usage of content and coupons from local retailers and restaurants, and preferred pricing for the Easy Ride service.

Figure: 9 Nissan easy ride self-driving taxi

9) Volvo-Autoliv-Ericsson-Zenuity:
This group is developing an end-to-end autonomous platform that integrates in-vehicle software and systems with connected safety data from other vehicles and infrastructure.
         They also recently launched an innovation lab for easier collaboration with start-ups.
         Volvo decided to take on the skies with its 360c concept, which seeks to compete with airliners when it comes to short-haul travel.
         Volvo 360c concept, revealed today as part of the company’s vision for the future of travel- autonomous, electric, connected and safe.
         Without the need for a driver’s seat and a steering wheel, the interior of the 360c can be reconfigured to become a sleeping cabin, a mobile office, a living room or an entertainment space.
         It runs on electric power, although no technical specifications exist – for now, this is all a flight of fancy.
         This is Volvo after all, so safety has also been given a rethink. Autonomous technology is expected to be introduced gradually rather than overnight, which means that fully self-driving vehicles will share the road with other motorists and pedestrians.
         360s include additional driver-assistance systems such as collision warning with pedestrian and cyclist detection, and distance alert.
         On tap is adaptive cruise control, which uses radar to match speeds with slower traffic and can bring the car to a complete stop. Lane-keeping assist helps prevent you from drifting over lane lines.
         The radar keeps you fairly well behind the lead car, so be prepared for a few folks to jam themselves into your lane.
          The Volvo is quick to apply the brakes, even to a full stop, but it's always best to pay careful attention.
         The lane-keeping assist vibrates and slightly tugs on the steering wheel if you cross a lane marker without signalling.
Figure: 10 Volvo 360s concept car

Figure:11 Interior of Volvo 360c

10) PSA:
The French automaker group that includes both Peugeot and Citroën have just been given the authorization to commence autonomous driving tests on French public roads with "non-expert" drivers behind the wheel.
         Group PSA is developing a new electronic architecture to improve the safety of autonomous vehicles for both passengers and data.
         PSA Group says it is committed to developing Level 3 autonomous driving, which allows hands-off capability but requires the driver to take back control at any time.
         The system, which PSA says can process 10 times more data than the current architecture, will have data protection and cyber security features built in, allowing customers to perform some maintenance operations and gain access to new services without going to a dealer
         Redundancy and backup will be provided by dual electric systems and batteries, and sensors will use different technologies to verify the accuracy of information they receive.
Figure: 12 Peugeot 3008 level 2 self-driving car.
11) Tesla:
A decade after Elon Musk announced Tesla's first master plan; he has just come up with its sequel-a grand outline that has been called out for being "absurd," and "insane," while simultaneously "brilliant" and "magnificent."
         Tesla model S, X, 3 is currently offering Autopilot, an advanced driver assist mode that can self-steer, brake, and switch lanes. 
         Autopilot is an optional driver-assist system for Tesla cars that Tesla CEO Elon Musk has promised will eventually turn the electric cars into fully self-driving vehicles.
         When Autopilot is engaged, cars can self-steer, adjust speed, detect nearby obstacles, apply brakes, and park. The technology uses a combination of radar, cameras, ultrasonic sensors, and GPS.
         Tesla says the on board computers in its Autopilot-enabled cars released since October 2016 can support full self-driving capabilities, and that this functionality will be added via firmware updates over time, subject to regulatory approval.
                What can Tesla's Autopilot do?
         Traffic-Aware Cruise Control:
         Auto steer
         Auto Lane Change
         Side Collision Warning
         Auto Park
         Summon
      •    Other features include Automatic Emergency Braking and Steering, Automatic High Beams, Blind Spot Detection, and Speed Assist

               Which sensors does Tesla's Autopilot use?
     Forward-looking radar: The radar used by Autopilot can see up to 160m ahead of the car, through "sand, snow, fog
     Eight cameras: 4 forward facing camera, wide angle camera and rear cameras
   Sonar: A 360-degree, ultrasonic sonar detects obstacles in an eight-meter radius around the car. 
    GPS: A satellite navigation system can detect the car's position on the road.
        •      All Tesla vehicles--Model S, Model X, and the new Model 3--are equipped with a custom-version of the Nvidia Drive PX2 platform.
Figure:13 Tesla model S
                                 
                 Figure: 14 Tesla Model S Autopilot mode

Indian Scenario of self-driving car:
Global tech giants such as Google, Tesla giving major goals into bringing self-driving vehicles on road, India isn’t lagging much behind when it comes to efforts in automation technology. 

We have earlier seen companies like Tata willing to invest into this field, TCS developing self-driving Nano, Infosys bringing self-driving rickshaw, Tech Mahindra’s reported interest on working on autonomous driving system, and others, who are taking a step to change India’s perspective of self-driving technology on a global frontier. 

Here the list of such start-up companies in India:
         Flux Auto:
         This Bangalore based autonomous truck driving start-up converts/ transforms normal truck into a driverless truck.
          It aims to facilitate features like cruise control, lane keeping, collision avoidance, keeping a check on accidents, higher efficiency, amongst others in a much cost-effective manner.
          With an aim to democratize autonomous systems, the 16-member start-up is developing the autonomous system without using Lidar—an incredibly must-have component of most autonomous vehicles. 

         Fisheye box:
         Fisheye box, a tech start-up based out of Kolkata is making India a frontrunner in the global race of self-driving and has Autonomous Celerio to flaunt.
         The project Aero drive by the start-up is aimed at making relatively low-tech car loaded with cameras, sensors and antennae to present a modified version of the car to the world. With a budget of INR 20 Lakh.
         they are looking for investments to jump from a ‘proof of concept’ stage to the next level. Run by a joystick, the car is packed with voice recognition software that runs on a ‘Start engine’ voice command. 
         Fisheye box’s driverless car comes with AIOS, car operating system, technology innovation that has made its trail run.

Figure: 15 Autonomous Celerio


         Hi Tech Robotic System:
          Novus –Drive, a completely autonomous electric vehicle by the company was launched in the Auto Expo Motor show 2016 in India. 
         This completely autonomous car lets the passengers set destination on the map, check details of destination drop off location, use snooze button to the stop vehicle, have Cloud Based Intelligence for a fleet management system and audio/visual alerts in case of emergencies. 

         ATI Motors:
         ATI motors are into designing autonomous electric cargo vehicles, with a major focus on customers outside India. 
         With an interest to take the road ‘never taken before’, these founders came together to make new kind vehicle, which they call mule internally.

         Intradyne:
         Founded by Aneesh Agrawal, a former Qualcomm Inc. president for India and South Asia, and David Julian, a former principal engineer at the US-based chip-maker, Intradyne uses artificial intelligence to improve road and driver safety. 
         Based out of Bangalore and San Diego, US, Intradyne offers a powerful camera that analyses driving patterns and can help determine the cause of accident. 
         The device called Driver is the size of a soap bar and is attached to vehicle’s rear-view mirror and rests on the inside of windscreen, pointing towards road. 

         Saadat Robots:
         Making technologies like LIDAR and RADAR redundant, Saadat robots is developing on-and off roads self-driving technology that works in extremely difficult traffic scenarios and in unstructured environmental conditions such as in India. 
         The technology by this start-up enables self-driving vehicles to perceive their environments using off-the-shelf cameras.
         Working on GPS denied environments; it can produce high speed trajectories even in cluttered environments. He is currently using Mahindra Bolero to test the technology.

         Aura Robotic:
         Run by IIT Kharagpur alumni, Aura Robotics is working on autonomous shuttle.
         The electric shuttle by the team relies heavily on LIDAR, a sensor technology that uses laser light for localization and spotting anomalies. 
         The Aura Robotics team is currently working on two separate models of their shuttles, so that that depending on the needs of the area, vehicles can either follow closed loop route with predefined stops or can be called on-demand to pick and drop at user defined location.
         Y Combinator-backed Aura Robotics is currently testing their driverless shuttle system at several universities, and is actually beginning to deploy shuttles on the campus of Santa Clara University.

Figure: 16 AURO Autonomous shuttle

         Omnipresent Robot:
         India’s one of the leading robotics providers, apart from building drones for industrial inspections, it is working relentlessly in the area of driverless cars. 
         It uses a combination of ID Laser sensors, stereo cameras and combines their data through sensor fusion to generate 3D maps. 
          They also have plans to deploy monocular camera for detecting traffic lights and decoding road signs. 
         Sirica 1.0 from IIT Bombay:
         Autonomous ground vehicle by IIT Bombay’s Innovation Cell nailed the fourth spot globally in the Intelligent Ground Vehicle Competition (IGVC) 2016, beating other universities from across the globe. 
         The team is looking forward to use GPS/INS, LiDAR and stereo cameras to gather information about the immediate environment at the right range,



Comments

Post a Comment

Popular posts from this blog

Voice Recognition Using web camera and Google API - Raspberry pi 3B (python)