Matches in SemOpenAlex for { <https://semopenalex.org/work/W3217139610> ?p ?o ?g. }
- W3217139610 abstract "This paper introduces differentiable higher-order control barrier functions (CBF) that are end-to-end trainable together with learning systems. CBFs are usually overly conservative, while guaranteeing safety. Here, we address their conservativeness by softening their definitions using environmental dependencies without loosing safety guarantees, and embed them into differentiable quadratic programs. These novel safety layers, termed a BarrierNet, can be used in conjunction with any neural network-based controller, and can be trained by gradient descent. BarrierNet allows the safety constraints of a neural controller be adaptable to changing environments. We evaluate them on a series of control problems such as traffic merging and robot navigations in 2D and 3D space, and demonstrate their effectiveness compared to state-of-the-art approaches." @default.
- W3217139610 created "2021-12-06" @default.
- W3217139610 creator A5025684214 @default.
- W3217139610 creator A5035339773 @default.
- W3217139610 creator A5058590941 @default.
- W3217139610 creator A5066830185 @default.
- W3217139610 date "2021-11-22" @default.
- W3217139610 modified "2023-10-16" @default.
- W3217139610 title "BarrierNet: A Safety-Guaranteed Layer for Neural Networks" @default.
- W3217139610 cites W1498436455 @default.
- W3217139610 cites W1980569135 @default.
- W3217139610 cites W2064675550 @default.
- W3217139610 cites W2194775991 @default.
- W3217139610 cites W2468433498 @default.
- W3217139610 cites W2489231587 @default.
- W3217139610 cites W2798791976 @default.
- W3217139610 cites W2963148914 @default.
- W3217139610 cites W2963403868 @default.
- W3217139610 cites W2963755523 @default.
- W3217139610 cites W2963970238 @default.
- W3217139610 cites W2973005513 @default.
- W3217139610 cites W2977676795 @default.
- W3217139610 cites W2998402843 @default.
- W3217139610 cites W3010871414 @default.
- W3217139610 cites W3035702124 @default.
- W3217139610 cites W3035945418 @default.
- W3217139610 cites W3038180127 @default.
- W3217139610 cites W3046925685 @default.
- W3217139610 cites W3090132620 @default.
- W3217139610 cites W3095242670 @default.
- W3217139610 cites W3096541186 @default.
- W3217139610 cites W3104608980 @default.
- W3217139610 cites W3106002372 @default.
- W3217139610 cites W3117756863 @default.
- W3217139610 cites W3119862069 @default.
- W3217139610 cites W3120214687 @default.
- W3217139610 cites W3131411499 @default.
- W3217139610 cites W3142708584 @default.
- W3217139610 cites W3148528090 @default.
- W3217139610 cites W3153025913 @default.
- W3217139610 cites W3155061166 @default.
- W3217139610 cites W3163219368 @default.
- W3217139610 cites W3172248255 @default.
- W3217139610 cites W3173365748 @default.
- W3217139610 cites W3173600339 @default.
- W3217139610 cites W3175797397 @default.
- W3217139610 cites W3176662521 @default.
- W3217139610 cites W3185969654 @default.
- W3217139610 cites W3195647294 @default.
- W3217139610 cites W3207730978 @default.
- W3217139610 cites W3209457451 @default.
- W3217139610 doi "https://doi.org/10.48550/arxiv.2111.11277" @default.
- W3217139610 hasPublicationYear "2021" @default.
- W3217139610 type Work @default.
- W3217139610 sameAs 3217139610 @default.
- W3217139610 citedByCount "0" @default.
- W3217139610 crossrefType "posted-content" @default.
- W3217139610 hasAuthorship W3217139610A5025684214 @default.
- W3217139610 hasAuthorship W3217139610A5035339773 @default.
- W3217139610 hasAuthorship W3217139610A5058590941 @default.
- W3217139610 hasAuthorship W3217139610A5066830185 @default.
- W3217139610 hasBestOaLocation W32171396101 @default.
- W3217139610 hasConcept C105795698 @default.
- W3217139610 hasConcept C126255220 @default.
- W3217139610 hasConcept C129844170 @default.
- W3217139610 hasConcept C134306372 @default.
- W3217139610 hasConcept C153258448 @default.
- W3217139610 hasConcept C154945302 @default.
- W3217139610 hasConcept C202615002 @default.
- W3217139610 hasConcept C203479927 @default.
- W3217139610 hasConcept C2524010 @default.
- W3217139610 hasConcept C2775924081 @default.
- W3217139610 hasConcept C33923547 @default.
- W3217139610 hasConcept C41008148 @default.
- W3217139610 hasConcept C47446073 @default.
- W3217139610 hasConcept C50644808 @default.
- W3217139610 hasConcept C6557445 @default.
- W3217139610 hasConcept C72434380 @default.
- W3217139610 hasConcept C86803240 @default.
- W3217139610 hasConceptScore W3217139610C105795698 @default.
- W3217139610 hasConceptScore W3217139610C126255220 @default.
- W3217139610 hasConceptScore W3217139610C129844170 @default.
- W3217139610 hasConceptScore W3217139610C134306372 @default.
- W3217139610 hasConceptScore W3217139610C153258448 @default.
- W3217139610 hasConceptScore W3217139610C154945302 @default.
- W3217139610 hasConceptScore W3217139610C202615002 @default.
- W3217139610 hasConceptScore W3217139610C203479927 @default.
- W3217139610 hasConceptScore W3217139610C2524010 @default.
- W3217139610 hasConceptScore W3217139610C2775924081 @default.
- W3217139610 hasConceptScore W3217139610C33923547 @default.
- W3217139610 hasConceptScore W3217139610C41008148 @default.
- W3217139610 hasConceptScore W3217139610C47446073 @default.
- W3217139610 hasConceptScore W3217139610C50644808 @default.
- W3217139610 hasConceptScore W3217139610C6557445 @default.
- W3217139610 hasConceptScore W3217139610C72434380 @default.
- W3217139610 hasConceptScore W3217139610C86803240 @default.
- W3217139610 hasLocation W32171396101 @default.
- W3217139610 hasOpenAccess W3217139610 @default.
- W3217139610 hasPrimaryLocation W32171396101 @default.
- W3217139610 hasRelatedWork W1890962660 @default.