Introducing Animal-AI v4.0.0!
NOTE: We are moving away from Animal-AI v3.1.4.experimental. This is the foundational build towards a stable v5.0.0 Animal-AI build.
What's New?
i. New valenced objects: DecoyGoal, DecoyGoalBounce (see docs here). However, in short, these new objects have no reward value and have a grey colour (only size can be changed), which behave as a regular valenced object, such as GoodGoal.
ii. New UI elements/data presentation: Arena Index and Total Objects (see screenshot below). The purpose of this new UI is to inform the user (play mode or training) the current arena index (starts at 1, which means arena indexed at 0 in the yaml file, this equates to arena index 1 in the UI) and the total number of the arenas defined (i.e. arena 1 out of 5). In addition, the purpose of the total objects UI element is significant for the current arena, meaning it is respective of how many objects are spawned in the current arena index, and not a combined integer for the whole yaml file.
iii. A few (perhaps not so notable changes) are made in the type of text used to present the UI information, such as reward value and health. The old, legacy logic has been removed and a new, modern and more advanced text component is now used for all texts displayed in the UI. This has provided an extra 4FPS on average (impressive for the simple change I made in my opinion). In conclusion, all text elements are now sharper and have a new, fresh look in terms of colour and backgrounds.
iv. In a more technical regard, general script optimisation was implemented, particularly on the scripts TrainingArena.cs and PlayerControls.cs. These scripts are now more readable and efficient.
v. Minor changes: the colour of the health bar has been changed from BLUE to GREEN (100% health). The colour spectrum is GREEN -> YELLOW (50% health) -> RED (0% health). The valanced objects RIPENGOAL and DECAYGOAL used to have purple colour in the logic (transition values directly altered the colour of the object colour, with purple implemented at some stage) which now has been removed. There is only GREY -> YELLOW for RIPENGOAL and YELLOW -> GREY for DECAYGOAL.
For a complete look, see CHANGELOG.md.
Getting Started:
Check our Installation Guide (here) and/or our Launching Animal-AI Guide (here) for a more detailed look.
But in brief:
- Download Animal-AI OS build for your system from below
- Extract files anywhere you'd like on your system (we use Winrar to package application contents)
- Run .exe/.app file (depending on your system) in the extracted folder of Animal-AI
N.B: If you are using Linux, you may need to make the .exe file executable: Simply run this command in your terminal: chmod +x env/AnimalAI.x86_64. If you are using MacOS, you may also need to run this command: chmod -R 777 [applicationName].app in your MacOS terminal to unlock permissions for running the application. Edit: Please also make sure when you extract the folder, you move the files inside the sub-directory to its parent folder.
What's Changed
- [MAJOR] PR - Overhaul Repository Structure by @alhasacademy96 in #45
Full Changelog: v3.1.4...v4.0.0
Screenshots:
