Probabilistic object tracking using a range camera

We address the problem of tracking the 6-DoF pose of an object while it is being manipulated by a human or a robot. We use a dynamic Bayesian network to perform inference and compute a posterior distribution over the current object pose. Depending on whether a robot or a human manipulates the object, we employ a process model with or without knowledge of control inputs. Observations are obtained from a range camera. As opposed to previous object tracking methods, we explicitly model self-occlusions and occlusions from the environment, e. g, the human or robotic hand. This leads to a strongly non-linear observation model and additional dependencies in the Bayesian network. We employ a Rao-Blackwellised particle filter to compute an estimate of the object pose at every time step. In a set of experiments, we demonstrate the ability of our method to accurately and robustly track the object pose in real-time while it is being manipulated by a human or a robot.

Authors

• Manuel Wüthrich
• Peter Pastor
• Mrinal Kalakrishnan
• Jeannette Bohg
• Stefan Schaal

Keywords

• Computational modeling
• Robot sensing systems
• Cameras
• Noise
• Real-time systems
• Robustness
• Object Tracking
• Time Step
• Process Model
• Posterior Probability
• Nonlinear Model
• Control Input
• Particle Filter
• Object Pose
• Robotic Hand
• Current Pose
• Dynamic Bayesian Network
• Measurement Noise
• Point Cloud
• Kalman Filter
• Assumption Of Independence
• Depth Images
• Robotic Arm
• Heavy-tailed
• Pose Estimation
• Depth Measurements
• Unscented Kalman Filter
• Extended Kalman Filter
• Objective Probability
• Bayesian Filtering
• Occluded Objects
• Human Arm
• Pose Tracking
• Probabilistic Graphical Models
• Tactile Sensor
• Sampling-based Approach