106323 - ARTIFICIAL INTELLIGENCE

Subject type: Mandatory

Coordinator: Adso Fernández Baena

Trimester: Second term

Credits: 4

Documentation mostly in English. Language used in class: Catalan. Exercises and tests in Catalan and / or English. 

This subject aims to introduce undergraduate students in the field of artificial intelligence, and specifically computational behaviors, showing them the application of some of their techniques in the construction of video games. Issues such as movement-based behaviors, including path-finding, and a small range of decision-making mechanisms of a reactive nature are seen. Decision-making mechanisms of a deliberative nature (planning) are also addressed, but in less depth. The theoretical aspects are worked on, in an expository way, and their subsequent practical application, aimed at solving, often guided, small problems. Class sessions combine both aspects in order to achieve a good balance between them. The practices (compulsory) and the exercises of class and at home conform the evaluative model of the asignatura. 

This subject has methodological and digital resources to make possible its continuity in non-contact mode in the case of being necessary for reasons related to the Codid-19. In this way, the achievement of the same knowledge and skills that are specified in this teaching plan will be ensured. 

The Tecnocampus will make available to teachers and students the digital tools needed to carry out the course, as well as guides and recommendations that facilitate adaptation to the non-contact mode. 

At the end of the course students must be able to:

E6.1 Design the software architecture of a video game according to specifications

E6.4 Classify and describe the main behaviors of artificial intelligence in video games and exemplify them with references in the video games market.

E6.6 Develop 2D and 3D video games (or parts thereof) in high-level languages ​​on platforms and engines intended for this purpose.

The subject mostly uses these two methodologies: the master class and problem solving. 

All the theoretical concepts of the subject will be exposed in theory classes (large groups) of a masterful nature. The “non-masterful” part of the sessions will be devoted to problem solving and short activities. Laboratory sessions with an eminently practical nature will also be scheduled. 

Students must attend all classes with a laptop with the ability to run the appropriate software for the subject. Teachers will report on what this software is and how it can be obtained.

Topic 1. Introduction. AI and AI for games. Computational behaviors

Topic 2. Motion control: "Steering behaviors"

2.1 Representation of the kinematic state

2.2 Basic and derived behaviors: seek, arrive, wander, velocity matching, ...

2.3 Combination of behaviors. Flocking

Topic 3. Pathfinding: "Pathfinding"

3.1 Representation of space: graphs

3.2 The A star algorithm

Item 4. Decision making

4.1 State machines

4.2 Behavioral trees

4.3 Goal-oriented behaviors 

4.4 Other decision-making mechanisms

In order to gather evidence of the achievement of the expected learning outcomes, the following evaluative activities will be carried out:

A1. Practice of motor behaviors and state machines (Laboratory practice / group work evidence of learning outcomes E6.1 and E6.6)

A2. Pathfinding practice (Laboratory practice / group work evidence of learning outcomes E6.1 and E6.6)

A3. Practice of motor behaviors (Laboratory practice / group work evidence of learning outcomes E6.1 and E6.6)

The content of the practices will make special emphasis in what indicates the title although will be able to contain other own contents of the asignatura and his area. 

A4. Final Exam  (evidence of learning outcomes E6.1, E6.4 and E6.6)

The student will have to show his knowledge with regard to the theoretical appearances of the asignatura and do small developments of practical character. 

General criteria of the activities

- The teacher will present a statement for each activity and the evaluation and / or rubric criteria

- The teacher will inform of the dates and the format of delivery of the activity

The grade of each student will be calculated following the following percentages:

A (1,2,3). Laboratory practices / group work: 50% (1/3 50% each)

A4. Final Exam: 50%

Final Note = A (1,2,3) · 0.5 + A4 · 0.5

Considerations:

- A4> = 5 is required to pass the subject. If this grade does not reach 5 then she herself will be the final grade. 

- An activity not delivered or delivered late and without justification (court summons or medical matter) counts as a 0.

- It is the student's responsibility to prevent plagiarism in all its forms. In the case of detecting a plagiarism, regardless of its scope, in any activity, it will correspond to have a note of 0. In addition, the professor will communicate to the Head of Studies the situation so that it takes applicable measures in the matter of regime. sanctioning. In the context of this subject, plagiarism also means using and / or adapting code that has not been developed entirely individually (or within the group in the case of group activities). Facilitating the code that leads to plagiarism is also a form of plagiarism and will be treated the same way. In summary, we can say that the evaluation activities must be solved in a strictly non-collaborative way (in the case of group activities, collaboration cannot transcend the heart of the group). 

Recovery

- It is necessary to obtain a mark> = 5 in the final exam of recovery to pass the asignatura.

- The mark of the recovery exam will be applied to the A4 activity (and the formula Final Grade = A (1,2,3) · 0.5 + A4 · 0.5 will be applied again) 

- In case of passing the recovery (A (1,2,3) · 0.5 + A4 · 0.5> = 5) the maximum final mark of the subject will be 5 

Millington, I. (2019). AI for games. Boca Raton, FL: CRC Press, Taylor & Francis Group.

Buckland, M. (2009). Programming Game Ai by Example. Plano, Tx: Wordware Public.