Search for: All records

We propose a learning framework, named Multi-Coordinate Cost Balancing (MCCB), to address the problem of acquiring point-to-point movement skills from demonstrations. MCCB encodes demonstrations simultaneously in multiple differential coordinates that specify local geometric properties. MCCB generates reproductions by solving a convex optimization problem with a multi-coordinate cost function and linear constraints on the reproductions, such as initial, target, and via points. Further, since the relative importance of each coordinate system in the cost function might be unknown for a given skill, MCCB learns optimal weighting factors that balance the cost function. We demonstrate the effectiveness of MCCB via detailed experiments conducted on one handwriting dataset and three complex skill datasets. 

Learning from Demonstration (LfD) is a popular approach to endowing robots with skills without having to program them by hand. Typically, LfD relies on human demonstrations in clutter-free environments. This prevents the demonstrations from being affected by irrelevant objects, whose influence can obfuscate the true intention of the human or the constraints of the desired skill. However, it is unrealistic to assume that the robot's environment can always be restructured to remove clutter when capturing human demonstrations. To contend with this problem, we develop an importance weighted batch and incremental skill learning approach, building on a recent inference-based technique for skill representation and reproduction. Our approach reduces unwanted environmental influences on the learned skill, while still capturing the salient human behavior. We provide both batch and incremental versions of our approach and validate our algorithms on a 7-DOF JACO2 manipulator with reaching and placing skills. 

In this paper, we present Combined Learning from demonstration And Motion Planning (CLAMP) as an efficient approach to skill learning and generalizable skill reproduction. CLAMP combines the strengths of Learning from Demonstration (LfD) and motion planning into a unifying framework. We carry out probabilistic inference to find trajectories which are optimal with respect to a given skill and also feasible in different scenarios. We use factor graph optimization to speed up inference. To encode optimality, we provide a new probabilistic skill model based on a stochastic dynamical system. This skill model requires minimal parameter tuning to learn, is suitable to encode skill constraints, and allows efficient inference. Preliminary experimental results showing skill generalization over initial robot state and unforeseen obstacles are presented. 

One‐second in situ measurements of CO and CO₂mole fractions were made aboard the National Aeronautics and Space Administration DC‐8 aircraft during the 2016 KORUS‐AQ joint air quality and atmospheric chemistry field campaign in South Korea. The ratio of CO to CO₂enhancement is used to characterize regional combustion source signatures. Calculations of the ∆CO/∆CO₂ratio were made with a short duration rolling window (60 s), filtered by the coefficient of determination (R²), and plotted as distributions to characterize air masses measured from the aircraft during the campaign. The KORUS‐AQ sampling domain was divided into analysis regions to facilitate the analysis. Over Seoul, the boundary layer shows a low‐ratio signature in the ∆CO/∆CO₂ratios, with more than 50% of the correlated slopes in the boundary layer falling below 1% ∆CO/∆CO₂, and 80% of the slopes between 0% and 2% ∆CO/∆CO₂. However, this behavior changes to a larger ratio distribution at higher altitudes. The West Sea receptor region was divided into three analysis sectors, by meteorological regime, and used in conjunction with measurements collected over China during the KORUS‐AQ campaign time period to characterize the Chinese ∆CO/∆CO₂ratio signature. Chinese‐type emissions have a slope distribution that is shifted to higher ratios and broadened compared to measurements over Seoul, with the bulk of the measurements between 2% and 4% ∆CO/∆CO₂, with few negative slopes. The measured ratio trends over South Korea are consistent with inventoried CO and CO₂emissions.