scholarly journals Advanced soft robot modeling in ChainQueen

Robotica ◽  
2021 ◽  
pp. 1-31
Author(s):  
Andrew Spielberg ◽  
Tao Du ◽  
Yuanming Hu ◽  
Daniela Rus ◽  
Wojciech Matusik
Keyword(s):  
Application Programming Interface ◽  
Material Point ◽  
Soft Robotics ◽  
Soft Robot ◽  
Inference Control ◽  
Simulation And Optimization ◽  
Efficient Simulation ◽  
Application Programming ◽  
Robot Modeling ◽  
Programming Interface

Abstract We present extensions to ChainQueen, an open source, fully differentiable material point method simulator for soft robotics. Previous work established ChainQueen as a powerful tool for inference, control, and co-design for soft robotics. We detail enhancements to ChainQueen, allowing for more efficient simulation and optimization and expressive co-optimization over material properties and geometric parameters. We package our simulator extensions in an easy-to-use, modular application programming interface (API) with predefined observation models, controllers, actuators, optimizers, and geometric processing tools, making it simple to prototype complex experiments in 50 lines or fewer. We demonstrate the power of our simulator extensions in over nine simulated experiments.

scholarly journals Coupling of Inference and Access Controls to Ensure Privacy Protection

2021 ◽  
Vol 11 (5) ◽  
pp. 529-535
Author(s):  
Jihane El Mokhtari ◽  
Anas Abou El Kalam ◽  
Siham Benhaddou ◽  
Jean-Philippe Leroy
Keyword(s):  
Access Control ◽  
Privacy Protection ◽  
Multidimensional Analysis ◽  
Sensitive Data ◽  
Inference Control ◽  
External Data ◽  
Control Rules ◽  
Access Controls ◽  
Policy Information ◽  
History Of

This article is devoted to the topic of coupling access and inference controls into security policies. The coupling of these two mechanisms is necessary to strengthen the protection of the privacy of complex systems users. Although the PrivOrBAC access control model covers several privacy protection requirements, the risk of inferring sensitive data may exist. Indeed, the accumulation of several pieces of data to which access is authorized can create an inference. This work proposes an inference control mechanism implemented through multidimensional analysis. This analysis will take into account several elements such as the history of access to the data that may create an inference, as well as their influence on the inference. The idea is that this mechanism delivers metrics that reflect the level of risk. These measures will be considered in the access control rules and will participate in the refusal or authorization decision with or without obligation. This is how the coupling of access and inference controls will be applied. The implementation of this coupling will be done via the multidimensional OLAP databases which will be requested by the Policy Information Point, the gateway brick of XACML to the various external data sources, which will route the inference measurements to the decision-making point.

Inference control of open relational queries under closed-world semantics based on theorem proving

2017 ◽  
Vol 70 ◽  
pp. 32-47 ◽  
Author(s):  
Joachim Biskup ◽  
Martin Bring ◽  
Michael Bulinski
Keyword(s):  
Theorem Proving ◽  
Inference Control ◽  
Closed World
Sign in / Sign up

Export Citation Format

Share Document