University of Technology Sydney

41077 Data Driven and Intelligent Robotics

Warning: The information on this page is indicative. The subject outline for a particular session, location and mode of offering is the authoritative source of all information about the subject for that offering. Required texts, recommended texts and references in particular are likely to change. Students will be provided with a subject outline once they enrol in the subject.

Subject handbook information prior to 2024 is available in the Archives.

UTS: Information Technology: Computer Science
Credit points: 6 cp

Subject level:

Undergraduate

Result type: Grade and marks

Recommended studies:

Basics of statistics and probability, Python programming, Introduction to Data Analytics

Description

Intelligent robots are a disruptive technology poised to transform business and society. However, developing intelligent robot behaviours is different to traditional business applications. Intelligent robots are real-time distributed systems that must make complex real-time decisions autonomously using data collected from a wide range of sources such as sensors and the internet. To deal with this complexity, professionals must not only make sense of complexity, context and social norms in real-world scenarios, but translate such insights to algorithms suitable for autonomous use by a robot.

In this subject, students work in groups to implement intelligent robot behaviours to solve a challenging real-world problem. Through this exercise, students develop their expertise in data analytics and machine learning. They develop an appreciation for the complexity, technical challenges and ethical issues associated with developing real-time data-driven intelligent systems. They gain an ability to recognise and select state-of-the-art algorithms, methodologies, techniques, experimental tools and evaluation methods. Students who complete this subject are ready to be productive members and thought-leaders in teams building the next generation of intelligent systems to transform business and society.

Subject learning objectives (SLOs)

Upon successful completion of this subject students should be able to:

1. Compare, contrast and select commodity sensors and data-transformation pipelines to generate data suitable for applications of intelligent robotics. (D.1)
2. Analyse and communicate the human social phenomena as well as the legal and social implications of data-driven robotics involved in a specific real-world application to inform a professional audience. (B.1)
3. Design novel data-driven robotic applications using insights obtained through observation and interview of users. (C.1)
4. Achieve professional mastery of the tools used for the development of intelligent robotic applications and demonstrate the ability to translate published research or theoretical models into functioning robotic systems that integrate data-driven intelligent algorithms. (D.1)
5. Evaluate, using an appropriate methodology, the implemented data-driven intelligent behaviour to predict real-world performance and limitations. (C.1)

Course intended learning outcomes (CILOs)

This subject also contributes specifically to the development of the following Course Intended Learning Outcomes (CILOs):

  • Socially Responsible: FEIT graduates identify, engage, interpret and analyse stakeholder needs and cultural perspectives, establish priorities and goals, and identify constraints, uncertainties and risks (social, ethical, cultural, legislative, environmental, economics etc.) to define the system requirements. (B.1)
  • Design Oriented: FEIT graduates apply problem solving, design and decision-making methodologies to develop components, systems and processes to meet specified requirements. (C.1)
  • Technically Proficient: FEIT graduates apply abstraction, mathematics and discipline fundamentals, software, tools and techniques to evaluate, implement and operate systems. (D.1)

Teaching and learning strategies

In this subject, students work to design and implement a social robot to solve a complex and sophisticated challenge problem inspired by a real-world commercial application or by published research.

The subject provides three hours in-class activities in each week.

Students will combine theory and practice to explore the challenges of building intelligent robots that (1) are aware of the real users’ needs and interactions, (2) consider legislative, ethical and social implications, and (3) integrate available state-of-the-art machine learning algorithms.

We build this subject upon the following three interrelated learning activities:

  1. Understand Theory. Students will conduct independent study and actively participate in collaborative discussions around selected readings. They will find novel connections between topics for the design of socially acceptable, ethical, effective and efficient data-driven robotics applications. They will be engaged in in-depth conversations during class.
  2. Conduct Research. Students will formulate research hypotheses, identify a sound methodology to test the hypotheses and run trials. They will engage in the practice of research through supervised in-class activities, as well as through independent research conducted beyond the classroom.
  3. Apply Theory and Research. Students will be supported through coaching in open-ended laboratory sessions in which they apply the acquired theoretical and research skills to produce viable designs and implementations of robotic applications. Students will be assisted in in-class coding exercises, and be expected to develop code outside of class to validate a prototype.

The theoretical aspects are delivered as reports and scientific research papers provided online. This material is to be read prior to class and will be discussed during the class. Students and the teaching team will interact and collaborate to reason about underlying implications of the topics under discussion. Students will be further engaged by formulating research questions about social phenomena related to the discussed topic.

In-class activities will also be used to explore the practical components of intelligent robotics. Students will be allocated to groups of three to five members and will work together to follow a user-centred development methodology that integrates the topics covered by the provided material.

Assessed deliverables are (a) the design of a data-driven and intelligent robot application, (b) a working prototype of the robot application, (c) a written report and seminar.

Students will apply insights gained in the subject to develop a software system for an existing robot platform, to that reveals insights about the intelligent interaction between users and robots. The individual effort of each student will be evaluated by individual contributions to the project code and other deliverables.

Groups of students will work together to prepare a final report and short seminar. The report will include an exploration of concepts relevant to data-driven intelligent robotics and the interaction between such systems and human users. A peer assessment process will be used to assess individual contributions to the project report. The outcomes and finding of the group will be presented in a short seminar to peers. After the presentation, the audience and the teaching team will have the chance to ask students questions. The questions can involve practical aspects of their work or theoretical topics related to the developed application. The answers to the questions will contribute to the final assessment of the presentation, together with the level of clarity of the presentation.

This subject integrates design thinking skills, sound research methodologies and practical coding skills. These features are crucial for the development of future intelligent system professionals.

Content (topics)

  1. Introduction to tools for intelligent robot applications development;
  2. Robotics sensors, their limitations, format, usage and processing for data-driven applications;
  3. Introduction to user-centred design in the context of intelligent robot applications;
  4. Legislative, ethical and social implications of data-driven robotics applications;
  5. User engagement and people perception in intelligent robot applications;
  6. Technologies and languages for implementing robotics applications;
  7. Methodologies for the formulation of hypotheses relevant to human-robot interaction research and to the design of experiments able to test them.

Assessment

Assessment task 1: Design a data-driven and intelligent robotic application.

Intent:

To demonstrate an understanding of a design process and to prepare a viable plan for a prototype to be implemented as part of Assessment 2 and 3.

Students will apply Human-robot interaction best practices in the design of the system. Students will be able to synthesise and prioritise choices, especially in terms of their technical and time feasibility.

Objective(s):

This assessment task addresses the following subject learning objectives (SLOs):

1, 2 and 3

This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs):

B.1, C.1 and D.1

Type: Design/drawing/plan/sketch
Groupwork: Group, group assessed
Weight: 30%
Length:

No required word limit, however it is suggested that submissions are approximately five pages.

Assessment task 2: Implement a prototype data-driven and intelligent robot application.

Intent:

To demonstrate an ability to deliver data-driven intelligent robotic applications.

Students will translate their design specifications into functioning robotic behaviours. The student will acquire leadership and teamwork skills by working in a team to deliver professional code that makes use of data analytics, artificial intelligence and insights into data-driven intelligent robots.

Objective(s):

This assessment task addresses the following subject learning objectives (SLOs):

3 and 4

This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs):

C.1 and D.1

Type: Project
Groupwork: Group, individually assessed
Weight: 40%

Assessment task 3: Report and seminar on data-driven and intelligent robotics

Intent:

To demonstrate a comprehensive understanding of data-driven and intelligent robotics including an ability to make insightful connections between theory, practice and experimental results.

Students will demonstrate an ability to think as researchers by testing the limitations of the integrated machine learning algorithm and by formulating clear hypotheses related to human-robot interaction research. The student will acquire the ability to provide sound methodologies to evaluate the designed and implemented robotic application by considering legislative, ethical and social implications. The student will be able to clearly communicate the choices and outcomes to an audience.

Objective(s):

This assessment task addresses the following subject learning objectives (SLOs):

5

This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs):

C.1

Type: Presentation
Groupwork: Group, individually assessed
Weight: 30%
Length:

No more than 8 pages report and 5 minute presentation + 10 minutes questions.

Minimum requirements

In order to pass the subject, a student must achieve an overall mark of 50% or more.

Required texts

Weekly required readings will be made available on UTSOnline.

Recommended texts

  • Python programming fundamentals. Lee, Kent D. Springer, 2011 (Available online in library);
  • Learning robotics using Python: design, simulate, program, and prototype an interactive autonomous mobile robot from scratch with the help of Python, ROS, and Open-CV. Lentin, Joseph (Available online in library);
  • Experimental design from user studies to psychophysics. Cunningham, Douglas W (Douglas William). CRC Press, 2012 (Available online in library);
  • Thoughtful machine learning with Python: a test-driven approach. Kirk, Matthew. O'Reilly, 2017 (Available online in library)
  • Python for data analysis: data wrangling with Pandas, NumPy, and IPython. McKinney, Wes. O'Reilly Media, Inc., 2017 (Available online in library)
  • OpenCV for Python Developers. Crawford,Patrick. lynda.com 2017 (Available online in library, video tutorials)

References

  • Learning robotics using Python: design, simulate, program, and prototype an interactive autonomous mobile robot from scratch with the help of Python, ROS, and Open-CV. Lentin, Joseph (Available online in library)
  • Python programming fundamentals. Lee, Kent D. Springer, 2011 (Available online in library)
  • Thoughtful machine learning with Python: a test-driven approach. Kirk, Matthew. O'Reilly, 2017 (Available online in library)
  • Python for data analysis: data wrangling with Pandas, NumPy, and IPython. McKinney, Wes. O'Reilly Media, Inc., 2017 (Available online in library)
  • OpenCV for Python Developers. Crawford,Patrick. lynda.com 2017 (Available online in library, video tutorials)
  • Experimental design from user studies to psychophysics. Cunningham, Douglas W (Douglas William). CRC Press, 2012 (Available online in library)