Predictive robot programming

One of the main barriers to automating a particular task with a robot is the amount of time needed to program the robot. Decreasing the programming time would facilitate automation in domains previously off limits. In this paper, we present a novel method for leveraging the previous work of a user to decrease future programming time: predictive robot programming. The decrease in programming time is accomplished by predicting waypoints in future robot programs and automatically moving the manipulator end-effector to the predicted position. To this end, we develop algorithms that construct simple continuous-density hidden Markov models by a state-merging algorithm based on waypoints from prior robot programs. We then use these models to predict the waypoints in future robot programs. While the focus of this paper is the application of predictive robot programming, we also give an overview of the underlying algorithms used and present experimental results.

Authors

• Kevin R. Dixon
• Martin Strand 0002
• Pradeep K. Khosla

Keywords

• Robot programming
• Production
• Robotics and automation
• Hidden Markov models
• Packaging
• Manipulators
• Computer languages
• Merging
• Design automation
• Design optimization
• Present Experimental Results
• Training Set
• Learning Algorithms
• Transition Probabilities
• Active Users
• Kullback-Leibler
• Current Task
• Online Program
• Prior Observations
• Prediction Scheme
• Directed Acyclic Graph
• Confidence Value
• Physical Tasks
• Formal Proof
• Discrete Intervals
• Task State
• Large Training Set
• Multiset
• Arc Welding
• Forward-backward Algorithm
• Availability Of Training Data
• Expected Value
• Small Training Set