Robotic Behavior Prediction Using Hidden Markov Models
This work addresses predictive abilities for robots in collaborative, adversarial, or obstacle avoidance scenarios, but it appears incremental as it applies an existing method (HMMs) to robotic behavior prediction.
The paper tackled the problem of enabling robots to predict dynamic agents' actions and behaviors in real-time, using hidden Markov models (HMMs) to model unobservable states, with experimental results from realistic simulations.
There are many situations in which it would be beneficial for a robot to have predictive abilities similar to those of rational humans. Some of these situations include collaborative robots, robots in adversarial situations, and for dynamic obstacle avoidance. This paper presents an approach to modeling behaviors of dynamic agents in order to empower robots with the ability to predict the agent's actions and identify the behavior the agent is executing in real time. The method of behavior modeling implemented uses hidden Markov models (HMMs) to model the unobservable states of the dynamic agents. The background and theory of the behavior modeling is presented. Experimental results of realistic simulations of a robot predicting the behaviors and actions of a dynamic agent in a static environment are presented.