After following this module, the student will be able to: |
- To systematically approach a design project from user requirements to device evaluation. (synthesis and
evaluation, with a little bit of application and analysis)
- To design a robot for application to a biomedical problem using multidisciplinary knowledge from mechanical,
electrical, control and software engineering domains. (application and synthesis)
- To create a kinematic model of the robot to control the joints to perform useful movements and tasks. (analysis,
synthesis, application and evaluation)
- To extract biological signals from the human body that can be used to control a robot. (application and synthesis)
Individuals with movement disorders have difficulty to participate in daily life. Robots have the potential to assist them when needed, and in this module we will design and build a robot that does just that.|
Robotics is the branch of technology that deals with the design, fabrication, operation, and application of robots, as well as computer systems for their control, sensory feedback, and information processing. These technologies deal with automated machines that can take the place of humans in dangerous environments or manufacturing processes, or resemble humans in appearance, behavior, or cognition. Worldwide scientific and industrial demand for skilled engineers with advanced systems and control knowledge of robotic systems that can apply this knowledge in biomedical or general high-tech systems is strongly increasing.
The elective module BioRobotics applies high-tech systems &control knowledge of robotic design and fabrication to the biomedical interaction with the human body, and thereby combines a vast number of disciplines. During the module, a robot has to be built that interacts with the human body to improve the quality of life for the individual with a movement disorder. To enhance student motivation and participation, such an individual will be invited to participate in the project and to grade the final results.
Much of the interdisciplinary material and skills required in this module is new to most students, but with the help of an experienced and motivated staff, the results they have been achieving since 2013 are truly amazing.
In the project, students have to design and realize a robot. During this, they will learn to:
- Go through a design trajectory systematically, by analyzing impaired human function, specifying user
requirements and technical requirements, generating ideas and concepts, evaluating concepts using
modelling and calculation, presenting a final design, realizing the system in hard- and software, evaluating the
performance with human interaction, and reporting the results verbally and in writing.
- Integrate knowledge from multiple disciplines such as biomedical, mechanical, electrical, software and control
- Make mechatronic simulation models of two-dimensional robots, by which conceptual designs can be evaluated
on performance criteria such as precision, speed, stiffness, strength, play, friction, natural frequencies and
crossover frequency, on the basis of which the concepts can be adjusted.
- Obtain and process biological signals (EMG) for usage in steering a robot.
The project is chosen to maximize the application of the knowledge gained in the following courses:
- BioRobotics design project
- Control of Robotic Systems
- Robot Kinematics
- Biomedical Signal Analysis
- Programming of Embedded Systems (part of the Project)
BioRobotics Design Project:
In the project, students have to analyze the needs of the participating patient, build the mechanical construction of the robot using wood laser-cut to their specifications, to select motors and construction elements from specified catalogues, program the signal analysis and robot control methods in Python in an embedded controller, and analyze the performance and acceptability of the device when interacting with humans. Most of this will be new to all students, but with the help of an experienced and motivated staff, the results they achieve are truly amazing. The project combined with the four courses leads to a very efficient and lasting knowledge transfer.
Control of Robotic Systems:
The students will learn how to make a dynamical model of the robot and analyze its behavior in the time-domain and frequency domain. The students learn how to translate mechatronics system requirements to PID feedback controller design to control the dynamic behavior of the robot. The course will deal with methods to determine stability in both the continuous-time and discrete-time domain.
The students learn to apply geometrical concepts from Lie group theory to serial robotic manipulators; in this case to design and analyse planar robot kinematics. Derivation of direct forwards kinematics and forward/backward differential kinematics allow the students to implement high-level position control in their project's embedded control solution.
Biomedical Signal Analysis:
The student learns how to convert neurophysiological signals to useable control inputs for the robots. The signals are often highly non-linear and very noisy, and thus require extensive processing. Special attention is given to the time-frequency relation of signals, to be able to relate them to control theory of robotic systems.
In agreement with the TOM philosophy, the project and courses are strongly intertwined. All global learning objectives of the module are addressed through multiple educational forms, and therefore by multiple, complementary methods of assessment.
Programming of Embedded Systems:
The students learn how to program real-time software on an embedded platform in the Python programming language. Furthermore useful skills such as thinking about a program flow and the process of debugging will be taught to the students.
The global learning objectives are translated into course and project specific objectives:
- Course objectives are assessed using two multiple-choice (MC) exams during the first eight weeks (on the
Mondays in weeks 5 and 9).
- Project objectives are assessed through the evaluation of the design project outcomes (report, demonstration
and a video).
We use MC exams for multiple reasons. One, it allows us to test students just after they have seen and used the material in the first few weeks, to allow the teachers to modify the course materials when and where needed. This is especially useful as the entrance level of the students is hard to predict. Two, the time-separated interactions with the material (through the lectures, the MC exams, the project and the oral exam) is the best method to ensure retention.
The MC exams are given on Monday mornings in weeks 5 and 9 of the module, and are based on the material taught in the weeks before. We explicitly allow complicated questions that require mathematical analysis and give up to seven possible answers per questions. The final grades in the module are given per course, thus based on two times one-half of each MC exam.
One major benefit to organizing MC exams in the above manner has been the reducing in scheduled overlap between assessment, lectures and the project. By scheduling the exams on Monday morning, students can focus on the upcoming courses and project work by Monday afternoon. Furthermore, the final MC exam is scheduled for the Monday in the ninth week, leaving one full week to complete the design project without any interference from scheduled courses. Without the final exams, the final two weeks (weeks 9 and 10) are completely devoted to robot demonstrations and presentations, peer exchange of knowledge, and the oral exams.
Teachers and experts, using the following components, assess the learning objectives of the design project:
- PES Exam: 25% of project grade (individual)
- Movie Presentation: 25% of project grade (group)
- Live Demonstration: 25% of project grade (group)
- Report: 25% of project grade (group)
The tight integration of course and project objectives, in combination with the overlapping assessment methods using MC tests, written reports, movie presentation, and device demonstration, guarantees the validity of the final module grade, and that it reflects the true abilities of the individual students. The reliability of the grade is enhanced by the wide range of evaluators (teachers and domain experts), in combination with the large number of grades and the ability to see all grades in relation to each other by the module coordinator. Transparency is high as all grades are published as soon as they are available.
Assumed previous knowledge
|Introductory courses on statistics, dynamics and control theory. Some pre-knowledge of MATLAB is recommended, although self-study of MATLAB tutorials would be sufficient.|
|Bachelor Biomedical Engineering||Required materials|
|Lecture notes and online readers will be made available for free later.|
|Shiavi, "Introduction to Applied Statistical Signal Analysis", 3e editie, ISBN: 978-0-12-088581-7. Freely available online: http://www.sciencedirect.com/science/book/9780120885817|
|Self study without assistance|
|BioRobotics Design Project|
|Control of Robotic Systems|
|Biomedical Signal Analysis|