Stanford ASL
@StanfordASL
Followers
1K
Following
59
Media
17
Statuses
95
The Autonomous Systems Lab (ASL) develops methodologies for the analysis, design, and control of autonomous systems. @Stanford
Joined April 2018
πScalable and safe deployment of generative robot policies in the real world requires that we actively monitor their behavior and issue warnings when they are failing. Check out @agiachris and @RohanSinhaSU latest work on runtime monitoring for generative robot policies.
In real-world deployment settings, generative robot policies are bound to encounter out-of-distribution scenarios that cause them to fail in unexpected ways. We present Sentinel, a runtime monitor that detects unknown failures of generative policies at deployment time!π(1/6)
0
0
3
π‘For human-robot interaction, human preferences need to be captured at all levels of the robot planning stack: task, motion, and control! Check out Text2Interaction from Jakob and @agiachris
How can robots incorporate human preferences into their plans? Introducing Text2Interaction: a long-horizon, skill-based planner that meets human preferences at the task, motion, and control levels zero-shot using code writing LLMs. Project site: https://t.co/ZpTa7sP7Zn
0
0
1
Join our ITSC Tutorial on "Data-driven Methods for Network-level Coordination of AMoD Systems Across Scales" tomorrow!
π’ Attending ITSC? Don't miss our Tutorial: "Data-driven Methods for Network-level Coordination of AMoD Systems Across Scales"!
0
0
1
β How can we enable real-time reactive control with LLMs for dynamic robotic systems? At #RSS2024 we present AESOP: A 2-stage (π’π) reasoning framework that uses LLMs to increase closed-loop robot trustworthiness in OOD scenarios! Site: https://t.co/eY93Z83BgX π§΅(1/6)
1
17
79
Can we leverage Transformer models to boost trajectory generation for spacecraft rendezvous? In our recent @IEEEAeroConf paper, we introduce ARTπ¨(Autonomous Rendezvous Transformer) to solve complex trajectory optimization problems. Websiteπ https://t.co/dEFhDBILq4 A thread π
3
3
9
π How can we detect system-level reasoning failures to improve the robustness of robotic systems in safety-critical settings? We use LLMs as intelligent runtime monitors to reason over and identify potentially problematic elements in a scene! π§ https://t.co/73CiaItpi8
1
3
15
π’ Announcing the first @corl_conf workshop on Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-based Autonomy! #CoRL2023 π― How can we build reliable robotic autonomy for the real world? π
Short papers due 10/6/23 π https://t.co/PulDqjtm5d π§΅(1/4)
1
15
33
Excited to present "Learning Control-Oriented Dynamical Structure from Data" next week at #ICML2023! We enforce factorized structure in learned dynamics models to enable performant nonlinear control. Paper: https://t.co/f79wPtohz9 Code (w/ #JAX): https://t.co/jqorikwxt5
4
9
34
Excited to share that our paper on Graph-Reinforcement Learning was accepted at #ICML2023! We present a broadly applicable approach to solve graph-structured MDPs through the combination of RL and classical optimization. Website: https://t.co/qVAjiTgrRt π§΅π(1/n)+quoted tweet
Can we learn efficient algorithms to solve classical optimization problems over graphs? In our recent @LogConference paper, we propose graph reinforcement learning as a general framework to solve network control problems! π https://t.co/63dlfkKdsy π§΅π (1/n)
4
16
73
Exciting first day co-teaching @drmapavoneβs AA203: Optimal and Learning-Based Control, with @spenMrich at @StanfordEng! Interested in the intersections between optimal control and RL? Look no further, all course materials will be available at: https://t.co/hMNP0sEhbz
4
28
104
Out-of-distribution inputs derail predictions of ML models. How can we cope with OOD data in robotics? How do we even define what makes data OOD? We provide a perspective paper arguing a system-level view of OOD data in robotics! 𧡠(1/5) Now on Arxiv:
2
8
24
Can we learn efficient algorithms to solve classical optimization problems over graphs? In our recent @LogConference paper, we propose graph reinforcement learning as a general framework to solve network control problems! π https://t.co/63dlfkKdsy π§΅π (1/n)
1
25
128
New on arXiv: we present a learning control approach capable of safe and efficient online adaptation. Our approach combines elements of classical adaptive control, modern robust MPC, and Bayesian meta-learning to yield guaranteed-safe online adaptation! https://t.co/rxXzHMm06H π§΅
2
20
96
We won the best paper award at the AI4Space Workshop! Here's our framework for how ML models can *detect* and *adapt* to changing input distributions, using OOD detection, subsampling, and continual learning. https://t.co/29IEJYduq9
#eccv @drmapavone @StanfordASL @AerospaceCorp
0
4
3
Great collaboration between the Institute for Dynamic Systems and Control and the Automatic Control Lab at @ETH (Nicolas Lanzetti, @AndreaCensi, @EmilioFrazzoli) and the @StanfordASL at @Stanford (@drmapavone). Check out the early access version at:
0
1
3
We unify several SDP relaxations for ReLU neural network verification by providing an exact convex formulation. This provides a path for relaxations that systematically trade off tightness & efficiency https://t.co/OCH77v0Dnz w/ @robin_a_brown, Ed Schmerling & @MarcoPavoneSU 2/2
1
1
9
Can we verify the safety of a deep neural network for deployment in safety-critical settings? This is a non-convex problem in general, and there have been many existing relaxations constructed for it. 1/2
1
1
30
Former ASLer, Prof. @NavidAzizan has an opening for a postdoctoral scholar in his lab!
We are looking for a strong postdoctoral scholar in the foundations of machine learning and intelligent systems. More info: https://t.co/8GYqBa4CbN Please retweet and share with any strong candidates you may know!
0
0
1
Can we leverage tools from statistical inference to build safety critical warning systems with a guaranteed Ξ΅ false negative rate using as few as 1/Ξ΅ data points? Check out our new paper on sample-efficient safety assurances using conformal prediction!
2
0
4
Kicking off the new year with a paper accepted to #ICRA2022! In it, we propagate perceptual state uncertainty through trajectory forecasting, making use of a new statistical distance-based loss formulation to do so. Check it out on arXiv: https://t.co/AmZT0PjJYU. See you in May!
1
1
2