Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning

Toussaint, Marc; Allen, Kelsey R.; Smith, Kevin A.; Tenenbaum, Joshua B.

doi:10.15607/rss.2018.xiv.044

Cited by 238 publications

(249 citation statements)

References 26 publications

Supporting

Mentioning

249

Contrasting

Order By: Relevance

“…Robotic manipulation involving tools has been studied in the task and motion planning (TAMP) literature [24,25,19,48,18,32]. [45,8] propose to use logic programming together with known models to algorithmically discover tool-use. One challenge that limits the scalability of most logic-based systems and analytic model-based systems is that modeling errors quickly accumulate during execution, which often results in fragile system.…”

Section: Related Workmentioning

confidence: 99%

“…This understanding becomes especially useful when performing complex multi-object manipulation tasks, such as those involved in tool use: if a robot could predict how one object might interact with another, it would be able to autonomously construct tool-use behaviors on the fly. While fully-specified analytic and symbolic models of physics can allow fully observable systems to perform such tasks [45], acquiring such models is substantially more challenging when the environment can only be observed through image observations. Learning predictive models of low-level observations, such as camera image pixels, has a number of benefits.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improvisation through Physical Understanding: Using Novel Objects As Tools with Visual Foresight

Xie¹,

Ebert²,

Levine³

et al. 2019

Robotics: Science and Systems XV

View full text Add to dashboard Cite

Machine learning techniques have enabled robots to learn narrow, yet complex tasks and also perform broad, yet simple skills with a wide variety of objects. However, learning a model that can both perform complex tasks and generalize to previously unseen objects and goals remains a significant challenge. We study this challenge in the context of "improvisational" tool use: a robot is presented with novel objects and a user-specified goal (e.g., sweep some clutter into the dustpan), and must figure out, using only raw image observations, how to accomplish the goal using the available objects as tools. We approach this problem by training a model with both a visual and physical understanding of multi-object interactions, and develop a sampling-based optimizer that can leverage these interactions to accomplish tasks. We do so by combining diverse demonstration data with self-supervised interaction data, aiming to leverage the interaction data to build generalizable models and the demonstration data to guide the model-based RL planner to solve complex tasks. Our experiments show that our approach can solve a variety of complex tool use tasks from raw pixel inputs, outperforming both imitation learning and self-supervised learning individually. Furthermore, we show that the robot can perceive and use novel objects as tools, including objects that are not conventional tools, while also choosing dynamically to use or not use tools depending on whether or not they are required. Videos of the results are available online 1 .

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Improvisation through Physical Understanding: Using Novel Objects As Tools with Visual Foresight

Xie¹,

Ebert²,

Levine³

et al. 2019

Robotics: Science and Systems XV

View full text Add to dashboard Cite

show abstract

“…forces, positions, contact points, etc. ), the generation of reference signals by g(·) can be addressed in multiple ways, ranging from hand-tuned state machines [15,41], trajectory optimization [14,45], imitation learning [5,12,19,38], or reinforcement learning (RL) [9,23].…”

Section: Introductionmentioning

confidence: 99%

Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks

Martín-Martín

Lee

Gardner

et al. 2019

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

146

View full text Add to dashboard Cite

Reinforcement Learning (RL) of contact-rich manipulation tasks has yielded impressive results in recent years. While many studies in RL focus on varying the observation space or reward model, few efforts focused on the choice of action space (e.g. joint or end-effector space, position, velocity, etc.). However, studies in robot motion control indicate that choosing an action space that conforms to the characteristics of the task can simplify exploration and improve robustness to disturbances. This paper studies the effect of different action spaces in deep RL and advocates for variable impedance control in end-effector space (VICES) as an advantageous action space for constrained and contact-rich tasks. We evaluate multiple action spaces on three prototypical manipulation tasks: Path Following (task with no contact), Door Opening (task with kinematic constraints), and Surface Wiping (task with continuous contact). We show that VICES improves sample efficiency, maintains low energy consumption, and ensures safety across all three experimental setups. Further, RL policies learned with VICES can transfer across different robot models in simulation, and from simulation to real for the same robot. Further information is available at https://stanfordvl.github.io/vices.

show abstract

“…For example, Degrave et al [19] proposed to directly solve differentiable equations. Such systems have been deployed for manipulation and planning for tool use [20]. Battaglia et al [12] and Chang et al [13] have both studied learning object-based, differentiable neural simulators.…”

Section: B Differentiable Physical Simulatorsmentioning

confidence: 99%

Combining Physical Simulators and Object-Based Networks for Control

Ajay

Bauzá

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Physics engines play an important role in robot planning and control; however, many real-world control problems involve complex contact dynamics that cannot be characterized analytically. Most physics engines therefore employ approximations that lead to a loss in precision. In this paper, we propose a hybrid dynamics model, simulator-augmented interaction networks (SAIN), combining a physics engine with an object-based neural network for dynamics modeling. Compared with existing models that are purely analytical or purely data-driven, our hybrid model captures the dynamics of interacting objects in a more accurate and data-efficient manner. Experiments both in simulation and on a real robot suggest that it also leads to better performance when used in complex control tasks. Finally, we show that our model generalizes to novel environments with varying object shapes and materials.

show abstract

Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning

Cited by 238 publications

References 26 publications

Improvisation through Physical Understanding: Using Novel Objects As Tools with Visual Foresight

Improvisation through Physical Understanding: Using Novel Objects As Tools with Visual Foresight

Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks

Combining Physical Simulators and Object-Based Networks for Control

Contact Info

Product

Resources

About