HumDPM: a decision process model for modeling human-like behaviors in time-critical and uncertain situations

摘要: Generating human-like behaviors for virtual agents has become increasingly important in many applications, such as crowd simulation, training, digital entertainment, and safety planning. One of challenging issues behavior modeling is how make decisions given some time-critical uncertain situations. In this paper, we present HumDPM, a decision process model agents, which incorporates two factors human making situations: experience emotion. rather than relying on deliberate rational analysis, an agent makes its by matching past cases to the current situation. We propose detailed representation case investigate mechanisms situation assessment, execution. To incorporate emotion into introduce appraisal assessment elicitation. may be affected emotional states when: 1) deciding whether it necessary do re-match cases; 2) determining situational context; 3) selecting cases. illustrate effectiveness HumDPM simulation. A study typical scenarios conducted, shows varied composition leads different individual behaviors, due retrieval experiences variation agents' states.