Design and implementation of a teleoperator’s workstation
Cita com:
hdl:2117/389750
Document typeBachelor thesis
Date2023-06-21
Rights accessOpen Access
Except where otherwise noted, content on this work
is licensed under a Creative Commons license
:
Attribution-NonCommercial-NoDerivs 4.0 International
Abstract
The project aims to implement a way for a teleoperator to control an existing self-driving car if the autonomous driving algorithms fail to respond to the encountered situation. The project will rely on the existing code developed by the MechLab Team at the HTW in Dresden, who have converted a BMW i3 into a self-driving car using surround and proximity sensors and a homemade software that controls the vehicle's speed and steering. The car is also able to detect pedestrians and other obstacles thanks to a deep learning algorithm dedicated to this part. Teleoperation systems pose many challenges, such as providing the teleoperator with the same level of situational awareness as a driver in the car. The driver needs to focus more on the surroundings, and therefore teleoperated drivers will have to rest more often and take more breaks. To address this challenge, the teleoperation system will use high information density sensors, including LiDAR, radar, and ultrasonic sensors, to provide the driver with an overlay of detected obstacles and the predicted path, enhancing reality to compensate for latency in communication by taking some workload off the operator. Another big challenge is to switch between the autonomous and teleoperated driving modes, as there are different problems that can appear. Most noticeably, during the time it takes for the operator to get aware of the situation and respond to the call, the car must be able to safely stop and wait for instructions from the operator. The failure to do so could result in dangerous or even deadly situations for the autonomous vehicle’s occupants as well as for the other road users, who do not need to wait for the communication to be established. One of the last great challenges is allow stable and fast communication between the car and the teleoperator. This can be achieved by narrowing the data transmitted for example by reducing video quality in predefined cases, or by ensuring redundancy in the communication media. Nevertheless, a complete loss of communication is not impossible, so a protocol needs to be defined in order to safely halt the vehicle while waiting on the reconnection of the transmission. To fulfil this project, our team will use MATLAB and Simulink in combination with different toolboxes from the MathWorks company. We will try to develop a human-machine interface for the teleoperator, implement a way for the operator to take over control of the vehicle, build scenarios to test and simulate our different programs and much more. All of this is done in order to build safer and more reliable autonomous vehicles for the future.
Description
Treball desenvolupat en el marc del programa "European Project Semester".
DegreeMOBILITAT INCOMING
Files | Description | Size | Format | View |
---|---|---|---|---|
Project final report.pdf | Final report | 3,711Mb | View/Open | |
Technical report.pdf | Article | 548,8Kb | View/Open |