Industrial doctorate position open to improve safety in Robotic Systems in the Aerospace field
Are you interested in human-machine interaction? The EIT Doctoral School offers an industrial doctorate position at the Spanish Company SERTEC Engineering to investigate how to increase the safety of Robotic Systems in the Aerospace field. You will develop novel techniques to monitor behaviours of flexible robots based on the Robot Operating System (ROS) architecture.
Furthermore, the industrial doctorate will comprise the study of languages to express monitors that can analyse a global behaviour. Case studies will be built to assess the ability of the whole framework to improve the safety and security capabilities of state-of-the-art robots.
Even though, ROS individual components are programmed with high quality software, the interaction of the different nodes and components can cause malfunctioning, with potential safety consequences. The goal of the thesis is to develop a monitoring architecture for ROS based robots that allows the dynamical analyses of the global behaviour of the system including component interactions and human-machine interactions. This has to be pursued through a non-intrusive instrumentation and monitoring capabilities to be applied to existing robots and in a variety of workshops.
A typical robotic system, for example a KUKA KR90 R3100 EXTRA HA integrates several sensors like scanners, cameras, projectors, grippers, and magnetic sensors with actuators.
Robots have been used in aerospace industry as in others to take over repetitive tasks. There are many processes in the aerospace industry that are non-repetitive. A robot has to operate in different scenarios and interact with users and operators sharing the same workspace for several processes. This flexibility requirement faces complex challenges, compared to situations where robots perform repetitive tasks without human interaction. The human-machine interaction, needed in most of the processes in the aerospace industry, is prone to errors that can cause the robot to hang up, sensors generating wrong data, and worst of all, safety hazards for the humans involved.
Assuring safety in human robot interaction is a very important concern and it is a very challenging problem even with appropriate sensors and high-quality software, because potentially catastrophic situations can be caused by the "incorrect" interaction of human and machine and not by an error in the robot’s behaviour.
Several SERTEC developments have been implemented using the ROS. The ROS architecture is very flexible for developing robotic applications but this flexibility, especially in intercommunicating processes, leads to additional safety and security issues. During our developments, we have experienced many safety incidents, mainly due to instructions that are impossible to accomplish by the robots resulting in the activation of the robots' protection mechanisms halting the whole process. ROS also suffers from security issues related to message passing mechanisms because it is vulnerable to different kinds of attacks that become even more critical in robotic systems connected through cloud services.
The way this project will attack the problem of superimposition of behaviours using Maintainance Operation Protocol (MOP) is by building on top of the ROSRV infrastructure. The ROS architecture is based on messages. The software that controls and operates the different components of a robot communicates using messages. ROSRV uses a man-in-the-middle approach and intercepts the messages between the ROS nodes and the ROS Master and between ROS nodes that communicate with each other. In this manner, the ROSRV infrastructure allows to extract the information for the monitoring process, and to prevent undesired behaviours (a kind of enforcement) by preventing the transmission of messages.
The project will start from identifying an existing robot, with both in-house and third-party component software and equip it with a ROSRV infrastructure. SERTEC will also identify case studies and scenarios that are potentially threatening for safety and security. In a second phase, a monitor’s ad-hoc for simple scenarios, on top of the ROSRV infrastructure will be built, with the goal of preventing hazards. Then a powerful and flexible runtime verification language will be designed, including a monitoring algorithm and compiler that allows to tackle automatically the hand-crafter cases and more powerful scenarios. Finally, the proposal will be evaluated from the point of view of the effectiveness in terms of preventing pitfalls, and from the point of view of the efficiency in building MOP use for the case studies
The main result of this work will be a MOP infrastructure for safety and security of robots. This infrastructure will consist of the general architecture, a language to express properties, a compiler to create monitors from specifications and a runtime system build using ROS to run the monitoring process and impose behaviours on the robot. The second outcome of the PhD will be a collection of cases studies where behaviours are imposed on realistic robots preventing hazards and violations of security and safety situations.
The doctoral student involved in this programme will share its time between the Co-Location Centre of the EIT Digital Madrid Node, the premises of SERTEC and a three to six-month mobility in a European University/Research Centre/Company will be also part of the programme.
- Industrial partner: SERTEC Engineering
- Academic/research partner: IMDEA Software Institute / Universidad Politécnica de Madrid
- Number of available PhD positions: 1
- Duration: 4 years
- This PhD will be funded by EIT Digital and SERTEC Engineering
Those interested in applying should send an e-mail to firstname.lastname@example.org, including a CV, a motivation letter, and documents showing their academic track records.
Please apply before 28 September, 2018.