Project 1 (Mini Project on Vehicle Direction Detection

MBD Live class batch 2

# AIM

General Overview:

Identifying the direction of the vehicle is one of the important & diverse features in Autonomous driving & Advanced Driver Assistance Features. This particular sub-feature of identifying the direction of the vehicle is basically identifying the direction the vehicle is taking based on the camera input.

The camera reads the road signs & stores them into its memory with unique values for a left turn, right turn & straight drive. Depending on the direction it is taking, the final indication is given to the driver – as an indication if he is driving in the recommended direction or not.

Vehicle Direction Determination can also be coupled alongside features like GPS systems to identify whether the vehicle is reaching its destination in an optimized manner. This sub-feature can also be used along with Lane Detection, Highway Warning, Ramp Entry / Exit in Wrong Way Detection, etc.

The objective of Mini Project:

1. Development of MATLAB Simulink model as per requirement.
2. Tag the requirements to the Simulink model; tagging requirement 1 & requirement 2 to their corresponding subsystems is fine.
3. MBD compliant changes, Data Dictionary creation & code generation is added advantage, and that is not the scope of this project.
4. If choosing code generation, Storage class for Input signals: ImportedExtern; Storage class for Output signal: Export to File; Storage class for local signals: localizable; Storage class for calibration signals: Const.
5. Choose sample time for all signals as 0.01s

Requirement - 1:

• Steering wheel input as yaw rate (Signal name: SteeringWheel_YawDegreeInput) is the input for this system.
• This is compared against 3 angular values, one each for left turn, right turn & straight drive (Calibration Values: Right_Turn_AngularLimit, Left_Turn_AngularLimit, Straight_Drive_Steering_Angle) to say which specific direction the steering wheel is turning towards.
• Use switch blocks to compare & develop this requirement. Keep this requirement in a subsystem & the output of this requirement is a local signal (Signal Name: Vehicle_Turn_Status).

Requirement – 2:

• Keep this requirement as a separate subsystem. Inputs to this requirement are local signal from requirement 1 (Signal Name: Vehicle_Turn_Status) & an input signal from camera (Signal Name: CameraInput_RoadSign), which confirms the occurrence of a road sign.
• Signal Vehicle_Turn_Status is compared against calibration values (Calibration Values: RightTurn_RoadSign, LeftTurn_RoadSign, Straight_RoadSign), if each of them is found equal, then each of the three corresponding output is compared against the camera input signal,
• Using a logical operator block, only one among them is finally given as output signal (Signal Name: Vehicle_Direction_Indicator).

Signals & Calibration Data List:

Signal / Calibration Name	Signal Type	Data Type	Dimension	Min	Max	Initial Value	Units
SteeringWheel_YawDegreeInput	Input	Int16	1	-180	180	-	Deg
CameraInput_RoadSign	Input	Boolean	1	0	1	-	-
Vehicle_Turn_Status	Local	Int16	1	-180	180	-	Deg
Right_Turn_AngularLimit	Calibration	Int16	[1 1]	-180	180	30	Deg
Left_Turn_AngularLimit	Calibration	Int16	[1 1]	-180	180	-120	Deg
Straight_Drive_Steering_Angle	Calibration	Int16	[1 1]	-180	180	0	Deg
RightTurn_RoadSign	Calibration	Int16	[1 1]	-180	180	30
LeftTurn_RoadSign	Calibration	Int16	[1 1]	-180	180	-120
Straight_RoadSign	Calibration	Int16	[1 1]	-180	180	0
Vehicle_Direction_Indicator	Output	Boolean	1	0	1	-	-

# THEORY

The ultimate goal of ADAS feature development is to make our roads safer and better suited for fully autonomous vehicles in the long run. Still, manufacturers and buyers shouldn’t underestimate the importance of ADAS for meeting current automotive challenges. The most significant impact of advanced driver assistance systems is in providing drivers with essential information and automating difficult and repetitive tasks. This increases safety for everyone on the road.

ADAS vision systems and ADAS safety systems require lots of fused sensors to monitor the vehicle’s surroundings and what’s going on inside the car. The most commonly used ADAS sensors today are lidar, radar, and ultrasonic.

Software for self driving cars featuring ultrasonic sensors usually consists of multiple sensors located in the front and rear bumpers and side-view mirrors. They transmit short sound waves and measure the time it takes for them to travel to a target object and return to the receiver.

An ADAS safety system can rely on ultrasonic sensor technology for low-speed and short-range applications such as blind spot detection, self-parking, and parking assistance. Radar and lidar are both used by ADAS engineers for object detection, collision prevention, and interaction with traffic management systems.

Still, there are differences between these technologies. Lidar is the best solution for real-time detection, but it’s unpleasantly expensive for mass deployment. Radar sensors, especially long-range ones, are reliable enough and cheaper but lack precision when detecting small objects.

# SIMULATION

1.Design the Model

This the main model subsystem that having two inputs SteeringWheel_YawDegreeInput and CameraInput_RoadSign entering into the Vehicle Detection Determination subsystem and one output Vehicle_Direction_Indicator

Change the model settings to MBD compliance changes

This includes the solver settings, code generation target file and sample time for solver.

there are 2 requirements are designed in two subsystem

Requirement:- 1

• Steering wheel input as yaw rate (Signal name: SteeringWheel_YawDegreeInput) is the input for this system.
• This is compared against 3 angular values, one each for a left turn, right turn & straight drive (Calibration Values: Right_Turn_AngularLimit, Left_Turn_AngularLimit, Straight_Drive_Steering_Angle) to say which specific direction the steering wheel is turning towards.
• Use switch blocks to compare & develop this requirement. Keep this requirement in a subsystem & the output of this requirement is a local signal (Signal Name: Vehicle_Turn_Status).

Requirement:- 2

• Keep this requirement as a separate subsystem. Inputs to this requirement are a local signal from requirement 1 (Signal Name: Vehicle_Turn_Status) & an input signal from the camera (Signal Name: CameraInput_RoadSign), which confirms the occurrence of a road sign.
• Signal Vehicle_Turn_Status is compared against calibration values (Calibration Values: RightTurn_RoadSign, LeftTurn_RoadSign, Straight_RoadSign), if each of them is found equal, then each of the three corresponding output is compared against the camera input signal,
• Using a logical operator block, only one among them is finally given as output signal (Signal Name: Vehicle_Direction_Indicator).

2. Requirement Tagging

all the requirements are tagged in order for better understandability, anyone who is reviewing the model or who is trying to update the model can get a clear idea of what logic is used in the system. from the requirement manager, we can see what are all are tagged and the description

Requirement - 1:

Requirement – 2:

3. Developing SLDD

4. Signal Resolution for inputs and outputs

The inputs and output signals are resolved so the signals will take the values from sldd and pass them to the logical blocks. Otherwise, it may take some junk values if it is unresolved

5. Model Update.

The model update is done by Ctrl-D. this is done to check for any static errors. If there are no errors coming the model is working fine without any errors.

6. Simulation

The model is simulated too for checking for errors and warmings

7. Model Advisory Report

By running the model advisor tool, we can see commonly made modelling errors.

Model Advisor Report

9. Code generation

build model for generating code

Generated code

Result

The vehicle direction determination is modeled as per the requirement, all the signals are defined in the Simulink data dictionary. All input and output signals are resolved and propagated. Updated the model to check for static errors. Run the model advisor tool to see any errors. since there are no errors in Model Advisor Report. Code has been generated using Embedded Coder.