In computerized assessment of knowledge it is important to quickly estimate the competence state of a testee. This is particularly true for digital educational games where this kind of assessment has to be done in a non-invasive way, i.e., by avoiding any queries or interruptions. This paper presents the mathematical foundation of a model by which a large set of competence states is partitioned in disjoint subsets and the probability of a subject being in a particular state is calculated for the subsets of states and updated according to the monitored performance. Based on this calculation adaptive interventions can be automatically chosen by tailoring the upcoming problem to the competence state of the user.

Competence-based Knowledge Space Theory (CbKST) has been proven to be a very well-fitting basis for realizing personalization in technology-enhanced learning. Especially in the area of game-based learning, however, some extensions and improvements are needed.Personalization in a serious game cannot be regarded simply as the selection of game assets according to the individual learner's current competences but it must also pay heed to the up-keeping of a storyline, it must be ensured that no part of the story is omitted that may be necessary to understand a later part. Therefore, a CbKST-compatible Markovian model for storytelling is proposed.A second issue is the ongoing, non-invasive assessment of the learner's current competences during the game. Every action of the learner within the game should be taken into account for the competence assessment, and the assessment must be done in real-time, i.e. there must not be any delay caused by the assessment which would interrupt the flow of the game. A simplified update procedure for competence assessment within CbKST is suggested which can solve this issue, and simulation results are presented comparing the new procedure with the classical one.

Collecting ground truth-data for real-world applications is a non-trivial but very important task. In order to evaluate new algorithmic approaches or to benchmark system performance, they are inevitable. This is particularly true for robotics applications. In this paper we present our data collection for the biped humanoid robot Nao. Reflective markers were attached to Nao’s body, and the positions and orientation of its body and head were tracked in 6D with an accurate professional vision-based body motion tracking system. While doing so, the data of Nao’s internal state, i.e., the readings of all its servos, the inertial measurement unit, the force receptors plus a camera stream of the robot’s camera were stored for different, typical robotic soccer scenarios in the context of the RoboCup Standard Platform League. These data will be combined in order to compile an accurate ground-truth data set. We describe how the data were recorded, in which format they are stored, and show the usability of the logged data in some first experiments on the recorded data sets. The data sets will be made publicly available for the RoboCup’s Standard Platform League community.