#1

1.  SAE International has a formal classification system for autonomous vehicles, ranging from level 0 (not automated) to level 5 (fully automated). Briefly describe each of the levels.


Level 0
Human drivers control everything

Level 1
Assist the operation of some of the drivers under limited conditions such as maintaining a constant inter-vehicle distance, assisting the steering wheel during parking, and preventing deviation from the lane by steering wheel operation. The driver needs to control the car all the time including assisting work.

Level 2
By the simultaneous operation of many of driver assists, the vehicle itself can accelerate, decelerate, and handle the steering wheel. However, the driver always needs driving consciousness.

Level 3
Limited to specific situations such as highway driving, drivers can divert attention from driving. In other words, you do not have to look forward in limited circumstances. However, when necessary it asks the driver to respond immediately.

Level 4
Except for extreme weather conditions etc., the automatic driving system will control autonomous cars. An unexpected situation occurs and it can go or stop even in the absence of assistance from human drivers.

Level 5
The highest level is automatic operation.
Basically, the operation required by the driver is only the setting of the destination.
There are also cases where there are no parts such as steering wheels and pedals for operating a car.





2.  Use the internet to find at least two examples of autonomous vehicles that already exist. Which companies/universities/groups are they from? What autonomous behaviour do they exhibit (i.e., can parallel park itself, can drive itself)? How have they been tested?


Example 1

Honda's Legend
This was developed by Honda.
The test method is using a dummy doll etc. at a test run place inside Honda's facility.
The legend is equipped with a safe driving support system called Honda SENSING.
Honda sensing has eight functions and all Legends are installed.
The functions are as follows.

1 :Collision Mitigation Brake System(CMBS)
When there is a risk of collision with a vehicle ahead or a pedestrian.
First draw attention with display and sound in the display. In case of further approach, brakes activate.
In case of emergency, the brakes will be activated more strongly, to avoid collision avoidance and damage.
When there is a danger of colliding with an oncoming vehicle when it protrudes into the oncoming lane.
In addition to the display and sound in the display, warn with vibration of the steering wheel and encourage avoidance operation.
If the approach and avoidance action is not  timely, the brake is activated to avoid collision and damage.

2: Pedestrian accident reduction steering
Millimeter wave radar and monocular camera detect pedestrians and lanes on the roadside.
When a car goes out of the pedestrian lane and a collision with a pedestrian is predicted, it warns with the display and the sound in the display.
By further controlling the steering in the avoidance direction, urge the driver to avoid the operation.





3: Adaptive Cruise Control(ACC)
Inter-vehicle distance and speed difference between forward vehicle and millimeter wave radar and monocular camera are detected.
Accelerate / decelerate within the preset vehicle speed, follow the vehicle in front. It will control accelerator and brake so as to maintain appropriate inter-vehicle distance, and it will try to reduce driving load during highway driving. In Legend, Accord, we also have a function to keep you in line, when in traffic.

4: LaneKeepAssistSystem(LKAS)
Detect lane with monocular camera. We support the steering operation so that the vehicle maintains the center of the lane, and we will try to reduce the driving load on expressway etc. Also, if you are about to get out of the lane, you will also be warned by display on the display and vibration of the steering.

5: Outdoor departure suppression function
Detect lane with monocular camera. When the vehicle is about to depart from the lane, warning with display on the display and vibration of the steering. At the same time control the steering to return it back into the lane. Also, if it is predicted that the deviation amount is large, it will assist you not to deviate by brake control.

6: Miscarryback suppression function
Millimeter-wave radar detects that there are forward vehicles at short distance, while stopping and running at less than about 10 km / h.
It suppresses sudden acceleration when suddenly depressing the accelerator pedal and urges attention by the display and sound in the display.

7: Advance car departure notice functio
It will notify you by sound and indication when we continue to stop without noticing that the preceding car has started while waiting for a signal.

8: Label recognition function
Recognized road sign with monocular camera.
Prompt attention to signs by displaying on the display to support safe driving.





Example 2

Waymo's Chrysler Pacifica Hybrid minivans.
This was developed by Waymo.
Waymo is an autonomous car development company spun out of Google's parent company, Alphabet Inc. in December 2016. 
It Announced that it will collaborate with Honda this year
Tests are done on public roads everyday.
Also it is done in various places like,
Kirkland,WA, Metro Phoenix,AZ, Mountain View,CA, Austin, TX.
Waymo fully self-driving technology will handle all the driving so you can go from door to door without taking the wheel.


3.  Consider the SAE classifications from the question 1. At which level would each of the vehicles discussed in question 2 be? Provide rationale for your decisions.

Honda's legend is level two. Because this is driver assistance and driver can not distract attention.
Waymo is level 4. Because all the vehicle itself controls, but not in all cases.


4.  Please state at least 3 possible applications for autonomous vehicles, and why they would be beneficial for those applications.


Main causes of traffic accidents that are currently occurring are mostly due to mental and physical conditions of the driver such as mistakes in driver recognition, that is, driver's mistakes or overlooks, inattentive driving, sleeping and driving mistakes.
Automatic driving vehicles always recognize 360 degrees omnidirectional situation with sensors and cameras. It instantly makes decisions with computers and takes appropriate avoidance actions if it perceive the danger. 
It is certain that the reaction speed is overwhelmingly higher by computers than by humans.
Most of the traffic accidents may not happen if the automatic driving car is more widely used. 

It can be said that it is a particularly effective merit in developed countries where the aging population advances.

コメント

このブログの人気の投稿

#4

#2