Lei feng's network (search for "Lei feng's network", public interest): the author Zhu Yulong, auto industry engineers, automotive electronics.
In simple terms, is now Germany auto enterprises, took the United States and the world's technological corporations to come together to create a digital platform, coupled with Germany's chassis electronics and EE bus to innovation and evolution.
Recently, Intel and BMW Mobileye opened a small, automatic driving campaign group war. We should also address this in a future war of 100 years of automotive industry.
The tripartite cooperation will be based on the BMW Vision concept car Future Interaction I developing self-driving cars
The production version will be launched in 2021
Three parties built a better car platform, is a Union of Homo occupation
We briefly considered, in particular, to PK with AUDI BMW
1) field of ADAS
We seriously compare the two ADAS really has very little experience gap in L0~L1.
Sensor configuration such as Bosch, Conti and Kostal Dolce Gabbana iPhone Cases
77G Radar in addition to price convergence (you can choose)
ME+ camera module
2) field of automatic driving
According to the two concepts, BMW and AUDI for sensor is the same as the
IBEO LiDAR two to
We consider any different? What is worth to win?
BMW Group
1) BMW-led coordination
1.1 get should Radar&Lidar&Camera sensors
1.2 fix actuator
1.3 added V2X&Telematics
2) Intel computing platform
In May 2014, to the ZMP-funded $ 100 million, ZMP's "IZAC (Intel ZMP Autonomous Computer)", equipped with real time Linux box-shaped device. Processors Intel "Core I7". Will connect to the LIDAR, cameras, radar IZAC and vehicles and GPS sensor CAN bus connection, automatic application software running on Linux.
3) provided ME look around the image of deep learning
Here is Amnon Shashu inside a point I agree with: automatic support drove pillars
1. environmental technologies (Sensing): this piece of ME provided Visual approaches, and BMW is building a complete system redundancy.
Sensors perceive environmental data in road conditions by calculating the unit to establish environmental models
2. driving policy (Driving policy):
Automatic driving system need to learn how to obey the => training (DNN) and regulations (Rule based) Dolce & Gabbana iPhone 6 Cases
When to go
When to stop, and so on, need to
Systems need to be organized into systems to comply with the rules
3. mapping (Mapping)
Autopilot system needs to establish a very precise map, road conditions
Determine where
BMW redundancy system
Perceptual redundancy: sensor fusion
Map and redundant => path
High-precision map &REM redundancy
GPS+IMU position redundant
Compare AUDI, the patterns go a bit.
Secondary: Altera's Cyclone v SoC FPGAs: processing logic
Redundancy
32-bit TriCore Fail Safe
Core processing
NVIDIA Tegra K1: a lot of
Peripheral
MobilEye of Camera= EyeQ3: the handle in front of the camera, here are the supporting cast.
Ultrasound
77G radar
24G radar
Ultrasound
In a sense, success but also to take full advantage of the Visual system, which determines the cost and efficiency of the whole system
Two contrast, I believe that AUDI also needs more help to fight. In simple terms, is now Germany auto enterprises, took the United States and the world's technological corporations to come together to create a digital platform, coupled with Germany's chassis electronics and EE bus to innovation and evolution.
Summary:
1) Tesla's accident and more are a few waves along the way, behind great coalition war opened, and autopilot is not a company, now called ally
2) this is the ratio of inputs, is power, is also a long history of fighting
Lei Feng network Note: this paper, first author: automotive electronics design, please contact empower and retain the full information, not deletion of content.
No comments:
Post a Comment