skip to main content

Title: Learning compositional models of robot skills for task and motion planning
The objective of this work is to augment the basic abilities of a robot by learning to use sensorimotor primitives to solve complex long-horizon manipulation problems. This requires flexible generative planning that can combine primitive abilities in novel combinations and, thus, generalize across a wide variety of problems. In order to plan with primitive actions, we must have models of the actions: under what circumstances will executing this primitive successfully achieve some particular effect in the world? We use, and develop novel improvements to, state-of-the-art methods for active learning and sampling. We use Gaussian process methods for learning the constraints on skill effectiveness from small numbers of expensive-to-collect training examples. In addition, we develop efficient adaptive sampling methods for generating a comprehensive and diverse sequence of continuous candidate control parameter values (such as pouring waypoints for a cup) during planning. These values become end-effector goals for traditional motion planners that then solve for a full robot motion that performs the skill. By using learning and planning methods in conjunction, we take advantage of the strengths of each and plan for a wide variety of complex dynamic manipulation tasks. We demonstrate our approach in an integrated system, combining traditional robotics primitives more » with our newly learned models using an efficient robot task and motion planner. We evaluate our approach both in simulation and in the real world through measuring the quality of the selected primitive actions. Finally, we apply our integrated system to a variety of long-horizon simulated and real-world manipulation problems. « less
; ; ;
Award ID(s):
Publication Date:
Journal Name:
The International Journal of Robotics Research
Page Range or eLocation-ID:
866 to 894
Sponsoring Org:
National Science Foundation
More Like this
  1. Learning a robot motor skill from scratch is impractically slow; so much so that in practice, learning must typically be bootstrapped using human demonstration. However, relying on human demonstration necessarily degrades the autonomy of robots that must learn a wide variety of skills over their operational lifetimes. We propose using kinematic motion planning as a completely autonomous, sample efficient way to bootstrap motor skill learning for object manipulation. We demonstrate the use of motion planners to bootstrap motor skills in two complex object manipulation scenarios with different policy representations: opening a drawer with a dynamic movement primitive representation, and closingmore »a microwave door with a deep neural network policy. We also show how our method can bootstrap a motor skill for the challenging dynamic task of learning to hit a ball off a tee, where a kinematic plan based on treating the scene as static is insufficient to solve the task, but sufficient to bootstrap a more dynamic policy. In all three cases, our method is competitive with human-demonstrated initialization, and significantly outperforms starting with a random policy. This approach enables robots to to efficiently and autonomously learn motor policies for dynamic tasks without human demonstration.« less
  2. We present a novel method for performing integrated task and motion planning (TMP) by adapting any off-the-shelf sampling-based motion planning algorithm to simultaneously solve for a symbolically and geometrically feasible plan using a single motion planner invocation. The core insight of our technique is an embedding of symbolic state into continuous space, coupled with a novel means of automatically deriving a function guiding a planner to regions of continuous space where symbolic actions can be executed. Our technique makes few assumptions and offers a great degree of flexibility and generality compared to state of the art planners. We describe ourmore »technique and offer a proof of probabilistic completeness along with empirical evaluation of our technique on manipulation benchmark problems.« less
  3. In order to solve complex, long-horizon tasks, intelligent robots need to carry out high-level, abstract planning and reasoning in conjunction with motion planning. However, abstract models are typically lossy and plans or policies computed using them can be unexecutable. These problems are exacerbated in stochastic situations where the robot needs to reason about, and plan for multiple contingencies. We present a new approach for integrated task and motion planning in stochastic settings. In contrast to prior work in this direction, we show that our approach can effectively compute integrated task and motion policies whose branching structures encoding agent behaviors handlingmore »multiple execution-time contingencies. We prove that our algorithm is probabilistically complete and can compute feasible solution policies in an anytime fashion so that the probability of encountering an unresolved contingency decreases over time. Empirical results on a set of challenging problems show the utility and scope of our methods.« less
  4. We present a framework for planning complex motor actions such as pouring or scooping from arbitrary start states in cluttered real-world scenes. Traditional approaches to such tasks use dynamic motion primitives (DMPs) learned from human demonstrations. We enhance a recently proposed state of- the-art DMP technique capable of obstacle avoidance by including them within a novel hybrid framework. This complements DMPs with sampling-based motion planning algorithms, using the latter to explore the scene and reach promising regions from which a DMP can successfully complete the task. Experiments indicate that even obstacle-aware DMPs suffer in task success when used in scenariosmore »which largely differ from the trained demonstration in terms of the start, goal, and obstacles. Our hybrid approach significantly outperforms obstacle-aware DMPs by successfully completing tasks in cluttered scenes for a pouring task in simulation. We further demonstrate our method on a real robot for pouring and scooping tasks.« less
  5. There has been an explosion of ideas in soft robotics over the past decade, resulting in unprecedented opportunities for end effector design. Soft robot hands offer benefits of low-cost, compliance, and customized design, with the promise of dexterity and robustness. The space of opportunities is vast and exciting. However, new tools are needed to understand the capabilities of such manipulators and to facilitate manipulation planning with soft manipulators that exhibit free-form deformations. To address this challenge, we introduce a sampling based approach to discover and model continuous families of manipulations for soft robot hands. We give an overview of themore »soft foam robots in production in our lab and describe novel algorithms developed to characterize manipulation families for such robots. Our approach consists of sampling a space of manipulation actions, constructing Gaussian Mixture Model representations covering successful regions, and refining the results to create continuous successful regions representing the manipulation family. The space of manipulation actions is very high dimensional; we consider models with and without dimensionality reduction and provide a rigorous approach to compare models across different dimensions by comparing coverage of an unbiased test dataset in the full dimensional parameter space. Results show that some dimensionality reduction is typically useful in populating the models, but without our technique, the amount of dimensionality reduction to use is difficult to predict ahead of time and can depend on the hand and task. The models we produce can be used to plan and carry out successful, robust manipulation actions and to compare competing robot hand designs.« less