This paper presents a controller for the 2010 Simulated Car Racing Championship. The idea is not to create the fastest controller but a human-like controller. In order to achieve this, first we have created a process to build a model of the tracks while the car is running and then we used several neural networks which predict the trajectory the car should follow and the target speed. A scripted policy is used for the gear change and to follow the predicted trajectory with the predicted speed. The neural networks are trained with data retrieved from a human player, and are evaluated in a new track. The results shows an acceptable performance of the controller in unknown tracks, more than 20% slower than the human in the same tracks because of the mistakes made when the controller tries to follow the trajectory.

This paper is an initial approach to create a controller for the game TORCS by learning how another controller or humans play the game. We used data obtained from two controllers and from one human player. The first controller is the winner of the WCCI 2008 Simulated Car Racing Competition, and the second one is a hand coded controller that performs a complete lap in all tracks. First, each kind of controller is imitated separately, then a mix of the data is used to create new controllers. The imitation is performed by means of training a feed forward neural network with the data, using the backpropagation algorithm for learning.

This paper introduces the concept of using gaze as a sole modality for fully controlling player characters of fast-paced action computer games. A user experiment is devised to collect gaze and gameplay data from subjects playing a version of the popular Super Mario Bros platform game. The initial analysis shows that there is a rather limited grid around Mario where the efficient player focuses her attention the most while playing the game. The useful grid as we name it, projects the amount of meaningful visual information a designer should use towards creating successful player character controllers with the use of artificial intelligence for a platform game like Super Mario. Information about the eyes' position on the screen and the state of the game are utilized as inputs of an artificial neural network, which is trained to approximate which keyboard action is to be performed at each game step. Results yield a prediction accuracy of over 83% on unseen data samples and show promise towards the development of eye-controlled fast-paced platform games. Derived neural network players are intended to be used as assistive technology tools for the digital entertainment of people with motor disabilities.