Deep Sequential Models for Sampling-Based Planning

Kuo, Yen-Ling; Barbu, Andrei; Katz, Boris

doi:10.1109/iros.2018.8593947

Cited by 16 publications

(11 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent works [9][10] [11] [12] [13] [14] [15] have explored learning-based approaches to solving motion planning problems. While some employ a supervised imitation learning approach to train a model to mimic example paths, others use reinforcement learning techniques to learn path generation by trial and error.…”

Section: A Learning-based Planningmentioning

confidence: 99%

CLAMGen: Closed-Loop Arm Motion Generation via Multi-view Vision-Based RL

Akinola¹,

Wang²,

Allen³

2021

Preprint

View full text Add to dashboard Cite

We propose a vision-based reinforcement learning (RL) approach for closed-loop trajectory generation in an arm reaching problem. Arm trajectory generation is a fundamental robotics problem which entails finding collision-free paths to move the robot's body (e.g. arm) in order to satisfy a goal (e.g. place end-effector at a point). While classical methods typically require the model of the environment to solve a planning, search or optimization problem, learning-based approaches hold the promise of directly mapping from observations to robot actions. However, learning a collision-avoidance policy using RL remains a challenge for various reasons, including, but not limited to, partial observability, poor exploration, low sample efficiency, and learning instabilities. To address these challenges, we present a residual-RL method that leverages a greedy goalreaching RL policy as the base to improve exploration, and the base policy is augmented with residual state-action values and residual actions learned from images to avoid obstacles. Further more, we introduce novel learning objectives and techniques to improve 3D understanding from multiple image views and sample efficiency of our algorithm. Compared to RL baselines, our method achieves superior performance in terms of success rate.

show abstract

Section: A Learning-based Planningmentioning

confidence: 99%

CLAMGen: Closed-Loop Arm Motion Generation via Multi-view Vision-Based RL

Akinola¹,

Wang²,

Allen³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Here we describe how to augment them with networks that efficiently learn language, how to guide the planning process, and how to recognize when a plan described by a sentence has been completed. This is related to the approach of Kuo et al [7], DeRRT, which introduced deep sequential models for sampling-based planning. It took an RRT-based and described how to guide its behavior with a neural network.…”

Section: A Planning With Deep Rrtmentioning

confidence: 99%

“…Powerful models can control agents but do so from moment to moment without planning complex actions [1,2,3]. Planners on the other hand efficiently explore configuration spaces, often by building search trees [4,5], but require a target final configuration [6,7,8] or a symbolic specification of constraints [9,10].…”

Section: Introductionmentioning

confidence: 99%

Deep compositional robotic planners that follow natural language commands

Kuo¹,

Katz²,

Barbu³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

We demonstrate how a sampling-based robotic planner can be augmented to learn to understand a sequence of natural language commands in a continuous configuration space to move and manipulate objects. Our approach combines a deep network structured according to the parse of a complex command that includes objects, verbs, spatial relations, and attributes, with a sampling-based planner, RRT. A recurrent hierarchical deep network controls how the planner explores the environment, determines when a planned path is likely to achieve a goal, and estimates the confidence of each move to trade off exploitation and exploration between the network and the planner. Planners are designed to have near-optimal behavior when information about the task is missing, while networks learn to exploit observations which are available from the environment, making the two naturally complementary. Combining the two enables generalization to new maps, new kinds of obstacles, and more complex sentences that do not occur in the training set. Little data is required to train the model despite it jointly acquiring a CNN that extracts features from the environment as it learns the meanings of words. The model provides a level of interpretability through the use of attention maps allowing users to see its reasoning steps despite being an end-to-end model. This end-to-end model allows robots to learn to follow natural language commands in challenging continuous environments.

show abstract

“…Diankov and Kuffner [28] employs statistical techniques to sample around a search tree. Zucker et al [29], Kuo et al [30] formalize sampling as a model-free reinforcement learning problem and learn a parametric distribution. Since these problems are non i.i.d learning problems, they do require interactive learning and do not enjoy the strong guarantees of supervised learning.…”

Section: Related Workmentioning

confidence: 99%

LEGO: Leveraging Experience in Roadmap Generation for Sampling-Based Planning

Kumar

Mandalika

Choudhury

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

We consider the problem of leveraging prior experience to generate roadmaps in sampling-based motion planning. A desirable roadmap is one that is sparse, allowing for fast search, with nodes spread out at key locations such that a lowcost feasible path exists. An increasingly popular approach is to learn a distribution of nodes that would produce such a roadmap. State-of-the-art is to train a conditional variational auto-encoder (CVAE) on the prior dataset with the shortest paths as target input. While this is quite effective on many problems, we show it can fail in the face of complex obstacle configurations or mismatch between training and testing.We present an algorithm LEGO that addresses these issues by training the CVAE with target samples that satisfy two important criteria. Firstly, these samples belong only to bottleneck regions along near-optimal paths that are otherwise difficultto-sample with a uniform sampler. Secondly, these samples are spread out across diverse regions to maximize the likelihood of a feasible path existing. We formally define these properties and prove performance guarantees for LEGO. We extensively evaluate LEGO on a range of planning problems, including robot arm planning, and report significant gains over heuristics as well as learned baselines.

show abstract

Deep Sequential Models for Sampling-Based Planning

Cited by 16 publications

References 32 publications

CLAMGen: Closed-Loop Arm Motion Generation via Multi-view Vision-Based RL

CLAMGen: Closed-Loop Arm Motion Generation via Multi-view Vision-Based RL

Deep compositional robotic planners that follow natural language commands

LEGO: Leveraging Experience in Roadmap Generation for Sampling-Based Planning

Contact Info

Product

Resources

About