Claus Möbus

Probabilistic and Empirical Grounded Modeling of Agents in (Partial) Cooperative Traffic Scenarios

The Human Centered Design (HCD) of Partial Autonomous Driver Assistance Systems (PADAS) requires Digital Human Models (DHMs) of human control strategies for simulations of traffic scenarios. The scenarios can be regarded as problem situations with one or more (partial) cooperative problem solvers. According to their roles models can be descriptive or normative . We present new model architectures and applications and discuss the suitability of dynamic Bayesian networks as control models of traffic agents: Bayesian Autonomous Driver (BAD) models. Descriptive BAD models can be used for simulating human agents in conventional traffic scenarios with Between-Vehicle-Cooperation (BVC) and in new scenarios with In-Vehicle-Cooperation (IVC). Normative BAD models representing error free behavior of ideal human drivers (e.g. driving instructors) may be used in these new IVC scenarios as a first Bayesian approximation or prototype of a PADAS.

Further Steps towards Driver Modeling According to the Bayesian Programming Approach

The Human Centered Design (HCD) of Partial Autonomous Driver Assistance Systems (PADAS) requires Digital Human Models (DHMs) of human control strategies for simulating traffic scenarios. We describe first results to model lateral and longitudinal control behavior of drivers with simple dynamic Bayesian sensory-motor models according to the Bayesian Programming (BP) approach: Bayesian Autonomous Driver (BAD) models. BAD models are learnt from multivariate time series of driving episodes generated by single or groups of users. The variables of the time series describe phenomena and processes of perception, cognition, and action control of drivers. BAD models reconstruct the joint probability distribution (JPD) of those variables by a composition of conditional probability distributions (CPDs). The real-time control of virtual vehicles is achieved by inferring the appropriate actions under the evidence of sensory percepts with the help of the reconstructed JPD.

Representing semantic knowledge with 2-dimensional rules in the domain of functional programming

One of the many difficult problems in the development of intelligent computer aided instruction (ICAI) is the appropriate design of instructions and helps. This paper addresses the question of optimizing instructional and help material concerning the operational knowledge for the visual, functional programming language ABSYNT (ABstract SYNtax Trees). The ultimate goal of the project is to build a problem solving monitor (PSM) for this language and the corresponding programming environment. The PSM should analyse the blueprints of the students, give comments and proposals (SLEEMAN & HENDLEY, 1982). First, we will explain our motivation for choosing this domain of discourse. Second, we will shortly present the programming environment of ABSYNT. Third, we represent the development of two alternative 2-D-rulesets (appendix A, B), which describe the operational semantics of the ABSYNT interpreter. The development of the 2-D-rules was guided by cognitive psychology and cognitive engineering aspects and results of an empirical study. The study showed that the rules were comprehensible even for computer novices.

KARaCAs: knowledge acquisition with repertory grids and formal concept analysis for dialog system construction

We describe a new knowledge acquisition tool that enabled us to develop a dialog system recommending software design patterns by asking critical questions. This assistance system is based on interviews with experts. For the interviews we adopted the repertory grid method and integrated formal concept analysis. The repertory grid method stimulates the generation of common and differentiating attributes for a given set of objects. Using formal concept analysis we can control the repertory grid procedure, minimize the required expert judgements and build an abstraction based hierarchy of design patterns, even from the judgements of different experts. Based on the acquired knowledge we semi-automatically generate a Bayesian Belief Network (BBN), that is used to conduct dialogs with users to suggest a suitable design pattern for their individual problem situation. Integrating these different methods into our knowledge acquisition tool KARaCAs enables us to support the entire knowledge acquisition and engineering process. We used KARaCAs with three design pattern experts and derived approximately 130 attributes for 23 design patterns. Using formal concept analysis we merged the three lattices and condensed them to approximately 80 common attributes.

Towards the Theory-Guided Design of Help Systems for Programming and Modelling Tasks

This paper describes an approach to the design of online help for programming tasks and modelling tasks, based on a theoretical framework of problem solving and learning. The framework leads to several design principles which are important to the problem of when and how to supply help information to a learner who is constructing a solution to a given problem. We will describe two example domains where we apply these design principles: The ABSYNT problem solving monitor supports learners with help and proposals for functional programming. The PETRI-HELP system currently under development is intended to support the learning of modelling with Petri nets.

Toward the Design of Adaptive Instructions and Helps for Knowledge Communication with the Problem Solving Monitor ABSYNT

For approximately ten years computer aided knowledge communication disappeared from the research scene. Today, it has been reestablished under the abbreviations of ICAI (Intelligent Computer Aided Instruction) and ITS (Intelligent Tutoring Systems) with regular conferences, research journals and textbooks [1,2,3,4,5].

Tutors, Instructions and Helps

The goals of this paper are threefold. First we want to present a review of the literature on Computer assisted Instruction, second we want to discuss the quality of instructions in some texts and human-computer dialogs concerning Computer programming. Third we want to demonstrate the cognitive-science-based development of our programming environment ABSYNT. This includes the construction of iconic Instructions and helps which promise to be superior to verbal instractions and helps when properly designed.

Interactive Support for Planing Visual Programs in the Problem Solving Monitor ABSYNT: Giving Feedback to User Hypotheses on the Language Level

We try to demonstrate the improvement of intelligent computer-aided instruction (ICAI) by the development of an interactive help system, which checks hypotheses postulated by the user during the problem solving process. The system is capable to recognize even incomplete proposals and contains the knowledge to generate complete solutions of the programming tasks. Thus the interactive help system adaptively supports the planning activities of the user. This is done by a goals-means-relation (GMR) which contains the domain-knowledge to analyze and synthesize ABSYNT-programs. At present this knowledge is worked out for 37 tasks in our curriculum and is condensed into 462 rules. The complexity of the solution space is rather astonishing. The system is capable to recognize and generate several millions of solutions even if height of ABSYNT-trees is restricted to five nodes.

Instruction-Based Knowledge Acquisition and Modification: The Operational Knowledge for a Functional, Visual Programming Language*

Abstract This contribution deals with instruction-based knowledge acquisition in a fairly complex but well-defined domain. The domain is the operational knowledge about the interpreter of ABSYNT, a functional, visual programming language which was developed in our project. Runnable specifications of the ABSYNT-interpreter were translated into sets of visual rules, serving as instructional material for students to acquire the operational knowledge. We are concerned with the following questions: 1. 1. How do subjects acquire the operational knowledge while simulating the interpreter of ABSYNT with the help of the instructional material? 2. 2. How can the operational knowledge gained by subjects be described? For example, does this knowledge differ from the instructional material? If the mental representation of the operational knowledge is isomorphic to the instructional material, then hypotheses about certain performance aspects can be stated. An experiment was conducted in which dyades of programming novices acquired the computational knowledge for ABSYNT by computing the value of ABSYNT-programs with the help of the instructions, thus simulating the interpreter. The hypotheses were disconfirmed. The results suggest that the mental representation of the operational knowledge consists of larger units than the instructional material, leading to the following hypotheses about the acquisition process and the mental representation of the operational knowledge: 1. 1. When faced with a difficulty, there will be problem solving with the help of the instructions. Thus new knowledge is acquired by failure-driven learning. 2. 2. When faced with familiar situations, compound rules are built. Thus the existing knowledge is improved by success-driven learning.

Predicting the focus of attention and deficits in situation awareness with a modular hierarchical Bayesian driver model

Situation Awareness (SA) is defined as the perception of elements in the environment within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future [1]. Lacking SA or having inadequate SA has been identified as one of the primary factors in accidents attributed to human error [2]. In this paper we present a probabilistic machine-learning-based approach for the real-time prediction of the focus of attention and deficits of SA using a Bayesian driver model as a driving monitor. This Bayesian driving monitor generates expectations conditional on the actions of the driver which are treated as evidence in the Bayesian driver model.