Incremental action recognition and generalizing motion generation based on goal-directed features

Gräve, Kathrin; Behnke, Sven

doi:10.1109/iros.2012.6386116

Cited by 12 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By modifying and recombining existing MPs, they are able to represent complex trajectories. Other approaches try to identify MPs using Bayesian Binning [6], Transition State Clustering [7], Conditional Random Fields [8,9], Gaussian Mixture Models [10][11][12], Hidden Markov Models [13,14] or Temporal Alignment for Control [15]. Changepoint methods can be used for decomposing a task in a two-step process.…”

Section: Related Workmentioning

confidence: 99%

Learning Sequential Force Interaction Skills

et al. 2020

View full text Add to dashboard Cite

Learning skills from kinesthetic demonstrations is a promising way of minimizing the gap between human manipulation abilities and those of robots. We propose an approach to learn sequential force interaction skills from such demonstrations. The demonstrations are decomposed into a set of movement primitives by inferring the underlying sequential structure of the task. The decomposition is based on a novel probability distribution which we call Directional Normal Distribution. The distribution allows infering the movement primitive’s composition, i.e., its coordinate frames, control variables and target coordinates from the demonstrations. In addition, it permits determining an appropriate number of movement primitives for a task via model selection. After finding the task’s composition, the system learns to sequence the resulting movement primitives in order to be able to reproduce the task on a real robot. We evaluate the approach on three different tasks, unscrewing a light bulb, box stacking and box flipping. All tasks are kinesthetically demonstrated and then reproduced on a Barrett WAM robot.

show abstract

Section: Related Workmentioning

confidence: 99%

Learning Sequential Force Interaction Skills

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Various studies have focused on learning motion primitives from manually segmented motions (Gräve and Behnke, 2012 ; Manschitz et al, 2015 ). Manschitz et al proposed a method to generate sequential skills by using motion primitives that are learned in a supervised manner.…”

Section: Related Workmentioning

confidence: 99%

Segmenting Continuous Motions with Hidden Semi-markov Models and Gaussian Processes

et al. 2017

View full text Add to dashboard Cite

Humans divide perceived continuous information into segments to facilitate recognition. For example, humans can segment speech waves into recognizable morphemes. Analogously, continuous motions are segmented into recognizable unit actions. People can divide continuous information into segments without using explicit segment points. This capacity for unsupervised segmentation is also useful for robots, because it enables them to flexibly learn languages, gestures, and actions. In this paper, we propose a Gaussian process-hidden semi-Markov model (GP-HSMM) that can divide continuous time series data into segments in an unsupervised manner. Our proposed method consists of a generative model based on the hidden semi-Markov model (HSMM), the emission distributions of which are Gaussian processes (GPs). Continuous time series data is generated by connecting segments generated by the GP. Segmentation can be achieved by using forward filtering-backward sampling to estimate the model's parameters, including the lengths and classes of the segments. In an experiment using the CMU motion capture dataset, we tested GP-HSMM with motion capture data containing simple exercise motions; the results of this experiment showed that the proposed GP-HSMM was comparable with other methods. We also conducted an experiment using karate motion capture data, which is more complex than exercise motion capture data; in this experiment, the segmentation accuracy of GP-HSMM was 0.92, which outperformed other methods.

show abstract

“…In our implementation, we use an iterative segmentation and classification algorithm that relies on Hidden Markov Models (HMM) to encode spatio-temporal features characterizing actions [15]. By optimizing the likelihood of a segment ending at a point as well as the likelihood of the succeeding segment starting at that point, motion boundaries are delineated reliably.…”

Section: Imitation Learningmentioning

confidence: 99%

“…To solve the task, the system is equipped with four primitive actions: grasping an object, displacing it, pushing it to a destination without lifting it, and retracting the hand. These primitive actions were trained using the approach described in our earlier work [15] and can be sequenced arbitrarily to create complex movements.…”

Section: A Experimental Setupmentioning

confidence: 99%

“…For our experiments, human demonstrations are simulated using a statistical model trained on collected motion capture data [15]. Using the learned motion primitive models, action sequences generated by our system are converted to trajectories and executed by a controller that computes their effects on objects and updates a simulated environment accordingly.…”

Section: B Task Definitionmentioning

confidence: 99%

See 1 more Smart Citation

Learning sequential tasks interactively from demonstrations and own experience

Gräve

Behnke

2013

2013 IEEE/RSJ International Conference on Intelligent Robots and Systems

Self Cite

View full text Add to dashboard Cite

Abstract-Deploying robots to our day-to-day life requires them to have the ability to learn from their environment in order to acquire new task knowledge and to flexibly adapt existing skills to various situations. For typical real-world tasks, it is not sufficient to endow robots with a set of primitive actions. Rather, they need to learn how to sequence these in order to achieve a desired effect on their environment. In this paper, we propose an intuitive learning method for a robot to acquire sequences of motions by combining learning from human demonstrations and reinforcement learning. In every situation, our approach treats both ways of learning as alternative control flows to optimally exploit their strengths without inheriting their shortcomings. Using a Gaussian Process approximation of the state-action sequence value function, our approach generalizes values observed from demonstrated and autonomously generated action sequences to unknown inputs. This approximation is based on a kernel we designed to account for different representations of tasks and action sequences as well as inputs of variable length. From the expected deviation of value estimates, we devise a greedy exploration policy following a Bayesian optimization criterion that quickly converges learning to promising action sequences while protecting the robot from sequences with unpredictable outcome. We demonstrate the ability of our approach to efficiently learn appropriate action sequences in various situations on a manipulation task involving stacked boxes.

show abstract

Incremental action recognition and generalizing motion generation based on goal-directed features

Cited by 12 publications

References 18 publications

Learning Sequential Force Interaction Skills

Learning Sequential Force Interaction Skills

Segmenting Continuous Motions with Hidden Semi-markov Models and Gaussian Processes

Learning sequential tasks interactively from demonstrations and own experience

Contact Info

Product

Resources

About