0 1 Briefly describe the basic components of ADCU (domain control)?
MCU, SOC, deserializer, CAN transceiver, network switch, power management module chip PMIC, etc.
0 2 What is the difference between MCU and SOC? What are the main application scenarios?
-
MCU is the main control chip, and the traditional MCU chip is mostly used for the main control chip in the domains that do not require high computing power, such as the chassis domain and the body domain;
-
SOC is a computing chip that integrates CPU, image processing GPU, audio processing DSP, deep learning acceleration unit NPU, memory and various I/O interfaces.
Considering that the intelligent cockpit domain and intelligent driving domain have extremely high requirements on computing power and security, the main control chip of the controller generally chooses the SOC chip to be used in conjunction with the MCU chip.
0 3 Briefly describe the implementation process of ADAS domain control software?
The software system of the controller is composed of the bottom operating system, the middle layer software and the upper layer application software.
Among them, the underlying operating system and middle layer software are mainly provided by domain controller manufacturers or software platform suppliers. The upper application layer has a higher degree of freedom, which can be developed by the OEM itself or commissioned by a third-party company.
Finally, the entire software layer is integrated and burned into the domain controller hardware. After adaptation and calibration with the entire vehicle, intelligent driving functions and scenarios can be realized on the vehicle side.
0 4 How is TTC calculated?
0 5 What are the radar output performance requirements?
1. The target ID should remain stable.
①The same target should only have one ID;
②When the target is within the FOV without occlusion interference, the same ID must be kept without jumping whether it changes from static to dynamic or dynamic to static.
2. The target position output should be consistent with the actual target movement, and there should be no abnormal situations such as position jumping and loss.
3. When the self-vehicle passes through obstacles such as vehicles parked on the roadside, walls, guardrails, flower beds, etc., the radar should not produce ghost images, and it is necessary to ensure stable and accurate tracking of other moving targets.
4. When the front radar detects a dynamic target, the distance error should not exceed ±0.4m in long-distance mode, ±0.1m in short-distance mode, and the speed error should not exceed ±0.1kph. For angle radar detection, the distance error is required to be no more than ±0.25m, and the speed error is no more than ±0.18kph.
5. The output error distance between the target center position and the actual target geometric center position does not exceed ±0.1m, and the long distance does not exceed ±0.25m.
0 6 When the forward-looking camera detects a dynamic target during the test, what is the measurement accuracy requirement of the acquisition device?
0 7 What are the common inhibitory conditions of AEB?
Excessive vehicle speed: When the vehicle speed exceeds a certain value, the AEB system may be inhibited to avoid sudden braking when driving at high speed, causing dangerous situations.
Obstacles too far away: AEB systems may be inhibited when obstacles are too far away from the vehicle, since braking is not necessary in this case.
Obstacles cannot be recognized: When the shape, size or color of the obstacle is not easy to recognize, the AEB system may be suppressed to avoid unnecessary braking measures if it is mistakenly recognized as an obstacle.
Sensor Failure: When a sensor of the AEB system fails or fails, the AEB system may inhibit.
Remarks: The suppression conditions of the AEB system are set by the OEM and may be different, so when using the AEB system, it should be tested according to the requirements.
0 8 What are the scenarios where AEB is falsely triggered?
0 9 What is the difference between MIL/SIL/HIL in the testing phase?
If the object under test is only an algorithm in the development stage, then only MIL/SIL can be used; if the object under test is a controller to be used in a real vehicle, a set of HIL equipment may be required to test in advance to find problems in advance.
communication
0 1 What are the two ros communication mechanisms?
-
Publish-subscribe mechanism between nodes;
-
Client-server request-response mechanism.
0 2 Which layer does the UDS belong to? How is the data transferred?
The UDS diagnostic service is located at the application layer, and the network layer is below the application layer, so the diagnostic data is pushed down from the application layer to the network layer, and the data is unpacked, packaged, and transmitted by the network layer.
CAN related
What are the commonly used CAN channels in 0 1 test? How are logical channels and physical channels mapped?
Example: set according to the unified standard.
0 2 How to connect CANoe, how to connect with ECU, what type of interface? Which ECUs need to be connected during the test?
ECU generally leads from the car, including chassis CAN, body CAN, FlexRay, radar, etc. to CANoe, and CANoe is connected to the computer. The interface is usually the head of DB9, or the connector drawn from the OBD on the car.
0 3 What is the difference between active safety and passive safety? What are the devices?
Passive safety refers to safety devices that can minimize personal injury after a traffic accident. Passive safety cannot prevent or avoid accidents, but they can minimize personal injury when an accident occurs. Common passive safety devices include airbags and seat belts.
Passive safety is not enough for automobile safety. In order to prevent/avoid/reduce the occurrence of safety accidents as far as possible,
A series of safety designs adopted on automobiles to prevent accidents are called active safety designs. Active safety is mainly to prevent problems before they happen, and work before accidents to reduce accidents. For example, the ABS anti-lock braking system prevents the car from locking and skidding; the ESP electronic stability system makes the car more stable when running and turning.
calibration
0 1 What is the difference between internal reference and external reference?
The internal reference refers to the inherent properties of the sensor itself, such as the focal length of the camera and lens distortion.
The external parameters are mainly used to unify the coordinate system between different sensors (Camera, Lidar, Radar) of the vehicle.
0 2 Do internal and external parameters need to be calibrated by themselves?
Usually only the external reference is required, and the internal reference is calibrated by the supplier.
data collection
0 1 How many cars does Datalogging have? How much mileage is collected in a single day? How long is the collection cycle?
Example: 3 collection vehicles, 300-400km/vehicle collection per day, and the collection period is about 3 months.
0 2 What scenes are covered by the collection?
-
Time period: day, evening, night;
-
Weather: sunny, cloudy, rainy and snowy;
-
Vehicles: large trucks, engineering vehicles, buses, cars, SUVs, pickups, motorcycles, work vehicles, accident vehicles, special-shaped vehicles;
-
VRU: pedestrians, children, cyclists;
-
Key scenarios: tunnels, bridges, ramps, large curvature bends, ramps, and elevated roads.
0 3 What is the difference between the side-view camera and the surround-view camera for collection?
Example: From a functional point of view, fisheye lenses are distributed in the four directions of the car body, front, rear, left, and right, and are mainly used for parking scenes.
The side view camera is divided into left front view/right front view/left rear view/right rear view, which are mainly used for driving assistance functions, such as DOW, LCA, etc.
Finally: The complete software testing video tutorial below has been organized and uploaded, and friends who need it can get it by themselves [Guaranteed 100% free]
Software Testing Interview Documentation
We must study to find a high-paying job. The following interview questions are the latest interview materials from first-tier Internet companies such as Ali, Tencent, and Byte, and some Byte bosses have given authoritative answers. Finish this set The interview materials believe that everyone can find a satisfactory job.
