Instrumentation and Control System for AI Based Driver Assistance System Onboard Datsun Car

The aim of the proposed project is to develop and deploy a low cost driver assistance system that would be capable to autonomously steer the steering wheel of an actual Datsun 120Y car. The far goal is to convert the Datsun car shown in Figure 1 into a driver assistance vehicle being able to complet

Project Title

Instrumentation and Control System for AI Based Driver Assistance System Onboard Datsun Car

Project Area of Specialization

Artificial Intelligence

Project Summary

The aim of the proposed project is to develop and deploy a low cost driver assistance system that would be capable to autonomously steer the steering wheel of an actual Datsun 120Y car. The far goal is to convert the Datsun car shown in Figure 1 into a driver assistance vehicle being able to completely take over the steering control function of the car using state-of-the-art computer vision and deep learning techniques. We are inspired by the methodology adopted by Tesla autonomous cars, which mainly rely on multiple cameras to achieve self driving capabilities. On the same lines, the driver assistance system that we are pitching in this proposal rely on the video feed from a dash cam, which will then be fed to a steering angle prediction algorithm to autonomously steer the car.

The team working on this project at HITEC University consists of five members. Due to limitation of Ignite NGIRI funding on the number of students, the overall project is divided into two projects:

Project 1: AI Based Driver Assistance Algorithm and Real-Time Testing for Datsun Car. (the other  project)

Project 2: Instrumentation and Control System for AI Based Driver Assistance System Onboard Datsun Car. (This Project)

About this project:

The team working on ‘Instrumentation and Control System for AI Based Driver Assistance System Onboard Datsun Car’ will deal with the hardware part ensuring successful interfacing of the AI Based Driver Assistance System (Project 1) with the car. Thus, the main contribution of this project will be limited to the necessary electronics and instrumentation. The intention is to make the Datsun car into a driver assistance vehicle able to completely take over the steering control function using state-of-the-art computer vision and deep learning techniques.  Figure 2 shows the basic workflow of this project. The input to the system will not be a sensor fusion, like in most modern-day expensive driver automation models, but rather it would be a simple visual input from a dash cam, which will be fed to the steering angle prediction algorithm for processing and subsequent maneuvering of the car.

The collaboration of both the projects (this project and Project 1) will result in hardware and software integrated system to control the steering wheel of the vehicle.

Project Objectives

Following are the main objectives of this project:

• Extracting real time values of steering angle.

An Optical Shaft Encoder (OSE) will be used to measure the angular movement of the steering wheel. The output of the OSE will be calibrated to measure the exact angular position of the steering wheel.

• Controlling the steering wheel electronically.

As shown in Figure 2, the measured steering angle will be given to the steering angle prediction algorithm running on Nvidia Jetson TX2 hardware board that will perform inference and predict the steering angle required to move the steering wheel. The objective is to design the required electromechanical hardware to move the steering wheel according to the command input received from the steering angle control algorithm. (The steering angle control algorithm will be developed by the team doing Project 1. Project 1 has also been submitted to NGIRI funding along with this project).

• PID controller design

A PID controller will be designed and implemented to stabilize the car’s steering wheel such that it follows the control input from the steering wheel prediction algorithm.

• To make a low cost yet efficient driver assistance model.

The present modern-day autonomous vehicle systems are expensive. One of the main objectives of this project is to design a low budget driver assistance system with minimal complexity.

Project Implementation Method

 In this section, the implementation aspect of this project will be discussed. Work on this project actually started in April 2020. The project team has achieved several milestones and is currently working to achieve the remaining targets that need to be completed within the stipulated time frame. This section describes the work that has been done and the remaining work to be completed.

Work done so far:

The installation an Optical Shaft Encoder (OSE) with the steering shaft has already been done using a pulley mechanism attached to the OSE on one end and to the steering shaft on the other end. The output of the OSE is given to a microcontroller based system that has been calibrated to find the current steering angle of the steering wheel. Figure 3 shows the OSE and pulley interfacing.

Figure 4 shows the picture of the electromechanical system that has been fitted to generate rotation in the gear train attached to it and in turn can rotate the steering wheel of the vehicle. The rotational control of the steering wheel can now be controlled electrically as well as manually using a mechanical lever. A sensor module will be further added to sense the driver’s hands such that if the driver is to take them off the steering wheel, the steering wheel control will go in the automatic control mode.

Work to be done:

The system up till now is able to move the steering electrically and estimate the current steering angle of the car. To further soften the steering movement, an Electronic Power Steering (EPS) is to be installed.  The final step would be to feed the current angle into a PID controller along with the predicted andrequired steering angle generated by the steering angle prediction software (done by Project 1). The stabilized output from the controller will be the input to the motor controlling the steering wheel. Thus, the software will have the control over the steering when needed. A solenoid valve system will also be incorporated to apply automatic breaking of the car so that in case of a sudden obstacles, the vehicle is able to stop, thus enabling the driver to avoid a dangerous situation.

Benefits of the Project

• This project can be a significant milestone towards Pakistan's contribution in developing autonomous driving system using state-of-the-art artificial intelligence technologies.
• This technology can be further developed to autonomous surveillance vehicles or border patrol vehicles.

Technical Details of Final Deliverable

Following are the deliverables with technical details of how they will be achieved.

Real time steering angle values:

This will be achieved by using an optical shaft encoder (OSE) mounted to the steering shaft using a pulley mechanism. The feed (pulses) from the OSE will be converted into angular position values.

Electrical control of  steering wheel:

To smoothen the steering, there is a need to install an electronic power steering (EPS).

Stabilizing the electrical inputs to the motor:

The output generated by the TX2 inference board will be given to the motor through a PID controller ensuring that the movement of the motor is smooth and follows the control command given by the TX2 inference board.

Interfacing hardware with software:

To interface the hardware developed in this project with the steering wheel control algorithm developed by team doing Project 1 ( ‘AI Based Driver Assistance Algorithm and Real-Time Testing for Datsun Car’).

Emergency breaking system:

To ensure security of the driver and the passenger, an emergency breaking system will come into action in case there is an object within close proximity of the car. The ‘AI Based Driver Assistance Algorithm and Real-Time Testing for Datsun Car’ project will give depth perception, and when the algorithm decides that an object is very close and there is a chance of collision, it will send a signal to the hardware which will be used to turn on a solenoid valve in order to apply emergency breaking.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

IT

Other Industries

Manufacturing , Transportation

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com