Search Results for author: Marco Pavone

Found 111 papers, 52 papers with code

Dolphins: Multimodal Language Model for Driving

no code implementations1 Dec 2023 Yingzi Ma, Yulong Cao, Jiachen Sun, Marco Pavone, Chaowei Xiao

The quest for fully autonomous vehicles (AVs) capable of navigating complex real-world scenarios with human-like understanding and responsiveness.

Autonomous Vehicles Language Modelling

Categorical Traffic Transformer: Interpretable and Diverse Behavior Prediction with Tokenized Latent

no code implementations30 Nov 2023 Yuxiao Chen, Sander Tonkens, Marco Pavone

Adept traffic models are critical to both planning and closed-loop simulation for autonomous vehicles (AV), and key design objectives include accuracy, diverse multimodal behaviors, interpretability, and downstream compatibility.

Autonomous Vehicles Common Sense Reasoning

A Language Agent for Autonomous Driving

1 code implementation17 Nov 2023 Jiageng Mao, Junjie Ye, Yuxi Qian, Marco Pavone, Yue Wang

Our approach, termed Agent-Driver, transforms the traditional autonomous driving pipeline by introducing a versatile tool library accessible via function calls, a cognitive memory of common sense and experiential knowledge for decision-making, and a reasoning engine capable of chain-of-thought reasoning, task planning, motion planning, and self-reflection.

Autonomous Driving Common Sense Reasoning +3

Real-time Control of Electric Autonomous Mobility-on-Demand Systems via Graph Reinforcement Learning

1 code implementation9 Nov 2023 Aaryan Singhal, Daniele Gammelli, Justin Luke, Karthik Gopalakrishnan, Dominik Helmreich, Marco Pavone

Operators of Electric Autonomous Mobility-on-Demand (E-AMoD) fleets need to make several real-time decisions such as matching available cars to ride requests, rebalancing idle cars to areas of high demand, and charging vehicles to ensure sufficient range.

Augmenting Lane Perception and Topology Understanding with Standard Definition Navigation Maps

no code implementations7 Nov 2023 Katie Z Luo, Xinshuo Weng, Yan Wang, Shuang Wu, Jie Li, Kilian Q Weinberger, Yue Wang, Marco Pavone

We propose a novel framework to integrate SD maps into online map prediction and propose a Transformer-based encoder, SD Map Encoder Representations from transFormers, to leverage priors in SD maps for the lane-topology prediction task.

Autonomous Driving Lane Detection

Interactive Joint Planning for Autonomous Vehicles

no code implementations27 Oct 2023 Yuxiao Chen, Sushant Veer, Peter Karkus, Marco Pavone

In particular, IJP jointly optimizes over the behavior of the ego and the surrounding agents and leverages deep-learned prediction models as prediction priors that the join trajectory optimization tries to stay close to.

Autonomous Vehicles Motion Planning +1

Closing the Loop on Runtime Monitors with Fallback-Safe MPC

no code implementations15 Sep 2023 Rohan Sinha, Edward Schmerling, Marco Pavone

When we rely on deep-learned models for robotic perception, we must recognize that these models may behave unreliably on inputs dissimilar from the training data, compromising the closed-loop system's safety.

Conformal Prediction

Robust Nonlinear Reduced-Order Model Predictive Control

no code implementations11 Sep 2023 John Irvin Alora, Luis A. Pabon, Johannes Köhler, Mattia Cenedese, Ed Schmerling, Melanie N. Zeilinger, George Haller, Marco Pavone

Real-world systems are often characterized by high-dimensional nonlinear dynamics, making them challenging to control in real time.

Dimensionality Reduction

Reinforcement Learning with Human Feedback for Realistic Traffic Simulation

no code implementations1 Sep 2023 Yulong Cao, Boris Ivanovic, Chaowei Xiao, Marco Pavone

This works aims to address this by developing a framework that employs reinforcement learning with human preference (RLHF) to enhance the realism of existing traffic models.


Language Conditioned Traffic Generation

1 code implementation16 Jul 2023 Shuhan Tan, Boris Ivanovic, Xinshuo Weng, Marco Pavone, Philipp Kraehenbuehl

In this work, we turn to language as a source of supervision for dynamic traffic scene generation.

Language Modelling Large Language Model +1

Multi-Predictor Fusion: Combining Learning-based and Rule-based Trajectory Predictors

no code implementations3 Jul 2023 Sushant Veer, Apoorva Sharma, Marco Pavone

Trajectory prediction modules are key enablers for safe and efficient planning of autonomous vehicles (AVs), particularly in highly interactive traffic scenarios.

Autonomous Vehicles Trajectory Prediction

Language-Guided Traffic Simulation via Scene-Level Diffusion

no code implementations10 Jun 2023 Ziyuan Zhong, Davis Rempe, Yuxiao Chen, Boris Ivanovic, Yulong Cao, Danfei Xu, Marco Pavone, Baishakhi Ray

Realistic and controllable traffic simulation is a core capability that is necessary to accelerate autonomous vehicle (AV) development.

Language Modelling Large Language Model

Bayesian Reparameterization of Reward-Conditioned Reinforcement Learning with Energy-based Models

no code implementations18 May 2023 Wenhao Ding, Tong Che, Ding Zhao, Marco Pavone

Recently, reward-conditioned reinforcement learning (RCRL) has gained popularity due to its simplicity, flexibility, and off-policy nature.

Offline RL reinforcement-learning

Graph Reinforcement Learning for Network Control via Bi-Level Optimization

1 code implementation16 May 2023 Daniele Gammelli, James Harrison, Kaidi Yang, Marco Pavone, Filipe Rodrigues, Francisco C. Pereira

Optimization problems over dynamic networks have been extensively studied and widely used in the past decades to formulate numerous real-world problems.


Partial-View Object View Synthesis via Filtered Inversion

no code implementations3 Apr 2023 Fan-Yun Sun, Jonathan Tremblay, Valts Blukis, Kevin Lin, Danfei Xu, Boris Ivanovic, Peter Karkus, Stan Birchfield, Dieter Fox, Ruohan Zhang, Yunzhu Li, Jiajun Wu, Marco Pavone, Nick Haber

At inference, given one or more views of a novel real-world object, FINV first finds a set of latent codes for the object by inverting the generative model from multiple initial seeds.

Exact Characterization of the Convex Hulls of Reachable Sets

1 code implementation30 Mar 2023 Thomas Lew, Riccardo Bonalli, Marco Pavone

We study the convex hulls of reachable sets of nonlinear systems with bounded disturbances.

Object Pose Estimation with Statistical Guarantees: Conformal Keypoint Detection and Geometric Uncertainty Propagation

1 code implementation CVPR 2023 Heng Yang, Marco Pavone

Geometric uncertainty propagation, on the other, propagates the geometric constraints on the keypoints to the 6D object pose, leading to a Pose UnceRtainty SEt (PURSE) that guarantees coverage of the groundtruth pose with the same probability.

Conformal Prediction Keypoint Detection

FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularization

2 code implementations CVPR 2023 Jiawei Yang, Marco Pavone, Yue Wang

One is to regularize the frequency range of NeRF's inputs, while the other is to penalize the near-camera density fields.

Neural Rendering Novel View Synthesis

Dynamic locational marginal emissions via implicit differentiation

no code implementations28 Feb 2023 Lucas Fuentes Valenzuela, Anthony Degleris, Abbas El Gamal, Marco Pavone, Ram Rajagopal

The method is model agnostic; it can compute LMEs for any convex optimization-based dispatch model, including some of the complex dispatch models employed by system operators in real electricity systems.

Learning Control-Oriented Dynamical Structure from Data

1 code implementation6 Feb 2023 Spencer M. Richards, Jean-Jacques Slotine, Navid Azizan, Marco Pavone

Even for known nonlinear dynamical systems, feedback controller synthesis is a difficult problem that often requires leveraging the particular structure of the dynamics to induce a stable closed-loop system.

A System-Level View on Out-of-Distribution Data in Robotics

no code implementations28 Dec 2022 Rohan Sinha, Apoorva Sharma, Somrita Banerjee, Thomas Lew, Rachel Luo, Spencer M. Richards, Yixiao Sun, Edward Schmerling, Marco Pavone

When testing conditions differ from those represented in training data, so-called out-of-distribution (OOD) inputs can mar the reliability of learned components in the modern robot autonomy stack.

Hybrid Multi-agent Deep Reinforcement Learning for Autonomous Mobility on Demand Systems

1 code implementation14 Dec 2022 Tobias Enders, James Harrison, Marco Pavone, Maximilian Schiffer

We consider the sequential decision-making problem of making proactive request assignment and rejection decisions for a profit-maximizing operator of an autonomous mobility on demand system.

Decision Making reinforcement-learning +1

DiffStack: A Differentiable and Modular Control Stack for Autonomous Vehicles

no code implementations13 Dec 2022 Peter Karkus, Boris Ivanovic, Shie Mannor, Marco Pavone

To enable the joint optimization of AV stacks while retaining modularity, we present DiffStack, a differentiable and modular stack for prediction, planning, and control.

Autonomous Vehicles

Adaptive Robust Model Predictive Control via Uncertainty Cancellation

no code implementations2 Dec 2022 Rohan Sinha, James Harrison, Spencer M. Richards, Marco Pavone

We propose a learning-based robust predictive control algorithm that compensates for significant uncertainty in the dynamics for a class of discrete-time systems that are nominally linear with an additive nonlinear component.


Online Distribution Shift Detection via Recency Prediction

no code implementations17 Nov 2022 Rachel Luo, Rohan Sinha, Yixiao Sun, Ali Hindy, Shengjia Zhao, Silvio Savarese, Edward Schmerling, Marco Pavone

When deploying modern machine learning-enabled robotic systems in high-stakes applications, detecting distribution shift is critical.

Expanding the Deployment Envelope of Behavior Prediction via Adaptive Meta-Learning

2 code implementations23 Sep 2022 Boris Ivanovic, James Harrison, Marco Pavone

Learning-based behavior prediction methods are increasingly being deployed in real-world autonomous systems, e. g., in fleets of self-driving vehicles, which are beginning to commercially operate in major cities across the world.

Meta-Learning regression

Data Lifecycle Management in Evolving Input Distributions for Learning-based Aerospace Applications

1 code implementation14 Sep 2022 Somrita Banerjee, Apoorva Sharma, Edward Schmerling, Max Spolaor, Michael Nemerouf, Marco Pavone

Algorithms within this framework are evaluated based on (1) model performance throughout mission lifetime and (2) cumulative costs associated with labeling and model retraining.

Active Learning Management +1

BITS: Bi-level Imitation for Traffic Simulation

1 code implementation26 Aug 2022 Danfei Xu, Yuxiao Chen, Boris Ivanovic, Marco Pavone

We empirically validate our method, named Bi-level Imitation for Traffic Simulation (BITS), with scenarios from two large-scale driving datasets and show that BITS achieves balanced traffic simulation performance in realism, diversity, and long-horizon stability.

Autonomous Vehicles

Robust Trajectory Prediction against Adversarial Attacks

no code implementations29 Jul 2022 Yulong Cao, Danfei Xu, Xinshuo Weng, Zhuoqing Mao, Anima Anandkumar, Chaowei Xiao, Marco Pavone

We demonstrate that our method is able to improve the performance by 46% on adversarial data and at the cost of only 3% performance degradation on clean data, compared to the model trained with clean data.

Autonomous Driving Data Augmentation +1

ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning

1 code implementation CVPR 2022 Yuxiao Chen, Boris Ivanovic, Marco Pavone

In this work, we present ScePT, a policy planning-based trajectory prediction model that generates accurate, scene-consistent trajectory predictions suitable for autonomous system motion planning.

Motion Planning Trajectory Prediction

Second-Order Sensitivity Analysis for Bilevel Optimization

3 code implementations4 May 2022 Robert Dyro, Edward Schmerling, Nikos Arechiga, Marco Pavone

Many existing approaches to bilevel optimization employ first-order sensitivity analysis, based on the implicit function theorem (IFT), for the lower problem to derive a gradient of the lower problem solution with respect to its parameters; this IFT gradient is then used in a first-order optimization method for the upper problem.

Bilevel Optimization

Safe Reinforcement Learning Using Black-Box Reachability Analysis

1 code implementation15 Apr 2022 Mahmoud Selim, Amr Alanwar, Shreyas Kousik, Grace Gao, Marco Pavone, Karl H. Johansson

Reinforcement learning (RL) is capable of sophisticated motion planning and control for robots in uncertain environments.

Motion Planning reinforcement-learning +2

Control-oriented meta-learning

1 code implementation14 Apr 2022 Spencer M. Richards, Navid Azizan, Jean-Jacques Slotine, Marco Pavone

Real-time adaptation is imperative to the control of robots operating in complex, dynamic environments.

Meta-Learning regression

Online Learning for Traffic Routing under Unknown Preferences

1 code implementation31 Mar 2022 Devansh Jalota, Karthik Gopalakrishnan, Navid Azizan, Ramesh Johari, Marco Pavone

at each period, we show that our approach obtains an expected regret and road capacity violation of $O(\sqrt{T})$, where $T$ is the number of periods over which tolls are updated.

Real-time Neural-MPC: Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms

2 code implementations15 Mar 2022 Tim Salzmann, Elia Kaufmann, Jon Arrizabalaga, Marco Pavone, Davide Scaramuzza, Markus Ryll

Our experiments, performed in simulation and the real world onboard a highly agile quadrotor platform, demonstrate the capabilities of the described system to run learned models with, previously infeasible, large modeling capacity using gradient-based online optimization MPC.

Motron: Multimodal Probabilistic Human Motion Forecasting

1 code implementation CVPR 2022 Tim Salzmann, Marco Pavone, Markus Ryll

We present Motron, a multimodal, probabilistic, graph-structured model, that captures human's multimodality using probabilistic methods while being able to output deterministic maximum-likelihood motions and corresponding confidence values for each mode.

Motion Forecasting

A Unified View of SDP-based Neural Network Verification through Completely Positive Programming

no code implementations6 Mar 2022 Robin Brown, Edward Schmerling, Navid Azizan, Marco Pavone

Verifying that input-output relationships of a neural network conform to prescribed operational specifications is a key enabler towards deploying these networks in safety-critical applications.

Graph Meta-Reinforcement Learning for Transferable Autonomous Mobility-on-Demand

1 code implementation15 Feb 2022 Daniele Gammelli, Kaidi Yang, James Harrison, Filipe Rodrigues, Francisco C. Pereira, Marco Pavone

Autonomous Mobility-on-Demand (AMoD) systems represent an attractive alternative to existing transportation paradigms, currently challenged by urbanization and increasing travel needs.

Meta Reinforcement Learning reinforcement-learning +1

Matching with Transfers under Distributional Constraints

no code implementations10 Feb 2022 Devansh Jalota, Michael Ostrovsky, Marco Pavone

To this end, we first consider the setting when the number of institutions (e. g., firms in a labor market) is one and show that equilibrium arrangements exist irrespective of the nature of the constraint structure or the agents' preferences.

Data-Driven Chance Constrained Control using Kernel Distribution Embeddings

no code implementations8 Feb 2022 Adam J. Thorpe, Thomas Lew, Meeko M. K. Oishi, Marco Pavone

We present a data-driven algorithm for efficiently computing stochastic control policies for general joint chance constrained optimal control problems.

A Simple and Efficient Sampling-based Algorithm for General Reachability Analysis

2 code implementations10 Dec 2021 Thomas Lew, Lucas Janson, Riccardo Bonalli, Marco Pavone

In this work, we analyze an efficient sampling-based algorithm for general-purpose reachability analysis, which remains a notoriously challenging problem with applications ranging from neural network verification to safety analysis of dynamical systems.

On the Problem of Reformulating Systems with Uncertain Dynamics as a Stochastic Differential Equation

no code implementations11 Nov 2021 Thomas Lew, Apoorva Sharma, James Harrison, Edward Schmerling, Marco Pavone

We identify an issue in recent approaches to learning-based control that reformulate systems with uncertain dynamics using a stochastic differential equation.

MTP: Multi-Hypothesis Tracking and Prediction for Reduced Error Propagation

1 code implementation18 Oct 2021 Xinshuo Weng, Boris Ivanovic, Marco Pavone

Recently, there has been tremendous progress in developing each individual module of the standard perception-planning robot autonomy pipeline, including detection, tracking, prediction of other agents' trajectories, and ego-agent trajectory planning.

Trajectory Planning

Coordinated Multi-Agent Pathfinding for Drones and Trucks over Road Networks

1 code implementation17 Oct 2021 Shushman Choudhury, Kiril Solovey, Mykel Kochenderfer, Marco Pavone

The second stage solves only for drones, by routing them over a composite of the road network and the transit network defined by truck paths from the first stage.

Multi-Agent Path Finding

Propagating State Uncertainty Through Trajectory Forecasting

1 code implementation7 Oct 2021 Boris Ivanovic, Yifeng Lin, Shubham Shrivastava, Punarjay Chakravarty, Marco Pavone

As a result, perceptual uncertainties are not propagated through forecasting and predictions are frequently overconfident.

Trajectory Forecasting

Injecting Planning-Awareness into Prediction and Detection Evaluation

1 code implementation7 Oct 2021 Boris Ivanovic, Marco Pavone

Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving.

Autonomous Driving Decision Making +1

Data Sharing and Compression for Cooperative Networked Control

1 code implementation NeurIPS 2021 Jiangnan Cheng, Marco Pavone, Sachin Katti, Sandeep Chinchali, Ao Tang

Sharing forecasts of network timeseries data, such as cellular or electricity load patterns, can improve independent control applications ranging from traffic scheduling to power generation.


Sample-Efficient Safety Assurances using Conformal Prediction

no code implementations28 Sep 2021 Rachel Luo, Shengjia Zhao, Jonathan Kuck, Boris Ivanovic, Silvio Savarese, Edward Schmerling, Marco Pavone

When deploying machine learning models in high-stakes robotics applications, the ability to detect unsafe situations is crucial.

Conformal Prediction Robotic Grasping

Tube-Certified Trajectory Tracking for Nonlinear Systems With Robust Control Contraction Metrics

1 code implementation9 Sep 2021 Pan Zhao, Arun Lakshmanan, Kasey Ackerman, Aditya Gahlawat, Marco Pavone, Naira Hovakimyan

This paper presents an approach towards guaranteed trajectory tracking for nonlinear control-affine systems subject to external disturbances based on robust control contraction metrics (CCM) that aims to minimize the $\mathcal L_\infty$ gain from the disturbances to nominal-actual trajectory deviations.

Motion Planning valid

On the Interplay between Self-Driving Cars and Public Transportation

no code implementations3 Sep 2021 Nicolas Lanzetti, Maximilian Schiffer, Michael Ostrovsky, Marco Pavone

Cities worldwide struggle with overloaded transportation systems and their externalities, such as traffic congestion and emissions.

Self-Driving Cars

Towards Data-Driven Synthesis of Autonomous Vehicle Safety Concepts

no code implementations30 Jul 2021 Karen Leung, Andrea Bajcsy, Edward Schmerling, Marco Pavone

As safety-critical autonomous vehicles (AVs) will soon become pervasive in our society, a number of safety concepts for trusted AV deployment have recently been proposed throughout industry and academia.

Autonomous Vehicles Inductive Bias

Bayesian Embeddings for Few-Shot Open World Recognition

no code implementations29 Jul 2021 John Willes, James Harrison, Ali Harakeh, Chelsea Finn, Marco Pavone, Steven Waslander

As autonomous decision-making agents move from narrow operating environments to unstructured worlds, learning systems must move from a closed-world formulation to an open-world and few-shot setting in which agents continuously learn new classes from small amounts of information.

Decision Making Few-Shot Learning

Rethinking Trajectory Forecasting Evaluation

no code implementations21 Jul 2021 Boris Ivanovic, Marco Pavone

Forecasting the behavior of other agents is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios with human-robot interaction, such as autonomous driving.

Autonomous Driving Decision Making +1

Joint Optimization of Autonomous Electric Vehicle Fleet Operations and Charging Station Siting

no code implementations1 Jul 2021 Justin Luke, Mauro Salazar, Ram Rajagopal, Marco Pavone

Charging infrastructure is the coupling link between power and transportation networks, thus determining charging station siting is necessary for planning of power and transportation systems.

Analysis and Control of Autonomous Mobility-on-Demand Systems

no code implementations28 Jun 2021 Gioele Zardini, Nicolas Lanzetti, Marco Pavone, Emilio Frazzoli

We provide a comprehensive review of methods and tools to model and solve problems related to autonomous mobility-on-demand systems.

Autonomous Vehicles

Graph Neural Network Reinforcement Learning for Autonomous Mobility-on-Demand Systems

1 code implementation23 Apr 2021 Daniele Gammelli, Kaidi Yang, James Harrison, Filipe Rodrigues, Francisco C. Pereira, Marco Pavone

Autonomous mobility-on-demand (AMoD) systems represent a rapidly developing mode of transportation wherein travel requests are dynamically handled by a coordinated fleet of robotic, self-driving vehicles.

Decision Making reinforcement-learning +1

Adaptive Robust Model Predictive Control with Matched and Unmatched Uncertainty

no code implementations16 Apr 2021 Rohan Sinha, James Harrison, Spencer M. Richards, Marco Pavone

We propose a learning-based robust predictive control algorithm that compensates for significant uncertainty in the dynamics for a class of discrete-time systems that are nominally linear with an additive nonlinear component.

Particle MPC for Uncertain and Learning-Based Control

no code implementations6 Apr 2021 Robert Dyro, James Harrison, Apoorva Sharma, Marco Pavone

As robotic systems move from highly structured environments to open worlds, incorporating uncertainty from dynamics learning or state estimation into the control pipeline is essential for robust performance.

Model-based Reinforcement Learning

Balancing Fairness and Efficiency in Traffic Routing via Interpolated Traffic Assignment

2 code implementations31 Mar 2021 Devansh Jalota, Kiril Solovey, Matthew Tsao, Stephen Zoepf, Marco Pavone

To address the inherent unfairness of SO routing, we study the ${\beta}$-fair SO problem whose goal is to minimize the total travel time while guaranteeing a ${\beta\geq 1}$ level of unfairness, which specifies the maximum possible ratio between the travel times of different users with shared origins and destinations.


Adaptive-Control-Oriented Meta-Learning for Nonlinear Systems

1 code implementation7 Mar 2021 Spencer M. Richards, Navid Azizan, Jean-Jacques Slotine, Marco Pavone

Real-time adaptation is imperative to the control of robots operating in complex, dynamic environments.

Meta-Learning regression

Sketching Curvature for Efficient Out-of-Distribution Detection for Deep Neural Networks

1 code implementation24 Feb 2021 Apoorva Sharma, Navid Azizan, Marco Pavone

In order to safely deploy Deep Neural Networks (DNNs) within the perception pipelines of real-time decision making systems, there is a need for safeguards that can detect out-of-training-distribution (OoD) inputs both efficiently and accurately.

Decision Making Out-of-Distribution Detection +1

Local Calibration: Metrics and Recalibration

no code implementations22 Feb 2021 Rachel Luo, Aadyot Bhatnagar, Yu Bai, Shengjia Zhao, Huan Wang, Caiming Xiong, Silvio Savarese, Stefano Ermon, Edward Schmerling, Marco Pavone

In this work, we propose the local calibration error (LCE) to span the gap between average and individual reliability.

Decision Making Fairness

Risk-sensitive safety analysis using Conditional Value-at-Risk

1 code implementation28 Jan 2021 Margaret P. Chapman, Riccardo Bonalli, Kevin M. Smith, Insoon Yang, Marco Pavone, Claire J. Tomlin

In addition, we propose a second definition for risk-sensitive safe sets and provide a tractable method for their estimation without using a parameter-dependent upper bound.

Sampling Training Data for Continual Learning Between Robots and the Cloud

no code implementations12 Dec 2020 Sandeep Chinchali, Evgenya Pergament, Manabu Nakanoya, Eyal Cidon, Edward Zhang, Dinesh Bharadia, Marco Pavone, Sachin Katti

Today's robotic fleets are increasingly measuring high-volume video and LIDAR sensory streams, which can be mined for valuable training data, such as rare scenes of road construction sites, to steadily improve robotic perception models.

Cloud Computing Continual Learning +2

Linear Reduced Order Model Predictive Control

1 code implementation6 Dec 2020 Joseph Lorenzetti, Andrew McClellan, Charbel Farhat, Marco Pavone

Model predictive controllers use dynamics models to solve constrained optimal control problems.

Leveraging Neural Network Gradients within Trajectory Optimization for Proactive Human-Robot Interactions

1 code implementation2 Dec 2020 Simon Schaefer, Karen Leung, Boris Ivanovic, Marco Pavone

To achieve seamless human-robot interactions, robots need to intimately reason about complex interaction dynamics and future human behaviors within their motion planning process.

Motion Planning Navigate +1

Task-relevant Representation Learning for Networked Robotic Perception

no code implementations6 Nov 2020 Manabu Nakanoya, Sandeep Chinchali, Alexandros Anemogiannis, Akul Datta, Sachin Katti, Marco Pavone

However, today's representations for sensory data are mostly designed for human, not robotic, perception and thus often waste precious compute or wireless network resources to transmit unimportant parts of a scene that are unnecessary for a high-level robotic task.

Motion Planning Representation Learning

Evidential Sparsification of Multimodal Latent Spaces in Conditional Variational Autoencoders

1 code implementation NeurIPS 2020 Masha Itkina, Boris Ivanovic, Ransalu Senanayake, Mykel J. Kochenderfer, Marco Pavone

Discrete latent spaces in variational autoencoders have been shown to effectively capture the data distribution for many real-world problems such as natural language understanding, human intent prediction, and visual scene representation.

Image Generation Motion Planning +1

CoCo: Learning Strategies for Online Mixed-Integer Control

no code implementations NeurIPS Workshop LMCA 2020 Abhishek Cauligi, Preston Culbertson, Mac Schwager, Bartolomeo Stellato, Marco Pavone

Mixed-integer convex programming (MICP) is a popular modeling framework for solving discrete and combinatorial optimization problems arising in various settings.

Combinatorial Optimization

Risk-Sensitive Sequential Action Control with Multi-Modal Human Trajectory Forecasting for Safe Crowd-Robot Interaction

no code implementations12 Sep 2020 Haruki Nishimura, Boris Ivanovic, Adrien Gaidon, Marco Pavone, Mac Schwager

This paper presents a novel online framework for safe crowd-robot interaction based on risk-sensitive stochastic optimal control, wherein the risk is modeled by the entropic risk measure.

Trajectory Forecasting

Safe Active Dynamics Learning and Control: A Sequential Exploration-Exploitation Framework

no code implementations26 Aug 2020 Thomas Lew, Apoorva Sharma, James Harrison, Andrew Bylard, Marco Pavone

In this work, we propose a practical and theoretically-justified approach to maintaining safety in the presence of dynamics uncertainty.

Meta-Learning Meta Reinforcement Learning

Multimodal Deep Generative Models for Trajectory Prediction: A Conditional Variational Autoencoder Approach

no code implementations10 Aug 2020 Boris Ivanovic, Karen Leung, Edward Schmerling, Marco Pavone

Human behavior prediction models enable robots to anticipate how humans may react to their actions, and hence are instrumental to devising safe and proactive robot planning algorithms.

Trajectory Prediction

Backpropagation through Signal Temporal Logic Specifications: Infusing Logical Structure into Gradient-Based Methods

1 code implementation31 Jul 2020 Karen Leung, Nikos Aréchiga, Marco Pavone

This paper presents a technique, named STLCG, to compute the quantitative semantics of Signal Temporal Logic (STL) formulas using computation graphs.

Continuous Meta-Learning without Tasks

2 code implementations NeurIPS 2020 James Harrison, Apoorva Sharma, Chelsea Finn, Marco Pavone

In this work, we enable the application of generic meta-learning algorithms to settings where this task segmentation is unavailable, such as continual online learning with a time-varying task.

Image Classification Meta-Learning +3

Efficient Large-Scale Multi-Drone Delivery Using Transit Networks

2 code implementations26 Sep 2019 Shushman Choudhury, Kiril Solovey, Mykel J. Kochenderfer, Marco Pavone

Our results show that the framework computes solutions typically within a few seconds on commodity hardware, and that drones travel up to $360 \%$ of their flight range with public transit.

Learning Stabilizable Nonlinear Dynamics with Contraction-Based Regularization

1 code implementation29 Jul 2019 Sumeet Singh, Spencer M. Richards, Vikas Sindhwani, Jean-Jacques E. Slotine, Marco Pavone

We propose a novel framework for learning stabilizable nonlinear dynamical systems for continuous control tasks in robotics.

Continuous Control

High-Dimensional Optimization in Adaptive Random Subspaces

no code implementations NeurIPS 2019 Jonathan Lacotte, Mert Pilanci, Marco Pavone

We propose a new randomized optimization method for high-dimensional problems which can be seen as a generalization of coordinate descent to random subspaces.

Vocal Bursts Intensity Prediction

On the Interaction between Autonomous Mobility on Demand Systems and Power Distribution Networks -- An Optimal Power Flow Approach

1 code implementation1 May 2019 Alvaro Estandia, Maximilian Schiffer, Federico Rossi, Justin Luke, Emre Can Kara, Ram Rajagopal, Marco Pavone

Specifically, we extend previous results on an optimization-based modeling approach for electric AMoD systems to jointly control an electric AMoD fleet and a series of PDNs, and analyze the benefit of coordination under load balancing constraints.

Self-Driving Cars

Network Offloading Policies for Cloud Robotics: a Learning-based Approach

no code implementations15 Feb 2019 Sandeep Chinchali, Apoorva Sharma, James Harrison, Amine Elhafsi, Daniel Kang, Evgenya Pergament, Eyal Cidon, Sachin Katti, Marco Pavone

In this paper, we formulate a novel Robot Offloading Problem --- how and when should robots offload sensing tasks, especially if they are uncertain, to improve accuracy while minimizing the cost of cloud communication?

Decision Making object-detection +1

Robust and Adaptive Planning under Model Uncertainty

no code implementations9 Jan 2019 Apoorva Sharma, James Harrison, Matthew Tsao, Marco Pavone

The first, RAMCP-F, converges to an optimal risk-sensitive policy without having to rebuild the search tree as the underlying belief over models is perturbed.

Decision Making

The Trajectron: Probabilistic Multi-Agent Trajectory Modeling With Dynamic Spatiotemporal Graphs

1 code implementation ICCV 2019 Boris Ivanovic, Marco Pavone

Developing safe human-robot interaction systems is a necessary step towards the widespread integration of autonomous agents in society.

Decision Making Motion Forecasting +2

Risk-Sensitive Generative Adversarial Imitation Learning

no code implementations13 Aug 2018 Jonathan Lacotte, Mohammad Ghavamzadeh, Yin-Lam Chow, Marco Pavone

We then derive two different versions of our RS-GAIL optimization problem that aim at matching the risk profiles of the agent and the expert w. r. t.

Imitation Learning

Meta-Learning Priors for Efficient Online Bayesian Regression

3 code implementations24 Jul 2018 James Harrison, Apoorva Sharma, Marco Pavone

However, this approach suffers from two main drawbacks: (1) it is computationally inefficient, as computation scales poorly with the number of samples; and (2) it can be data inefficient, as encoding prior knowledge that can aid the model through the choice of kernel and associated hyperparameters is often challenging and unintuitive.

Meta-Learning regression

BaRC: Backward Reachability Curriculum for Robotic Reinforcement Learning

1 code implementation16 Jun 2018 Boris Ivanovic, James Harrison, Apoorva Sharma, Mo Chen, Marco Pavone

Our Backward Reachability Curriculum (BaRC) begins policy training from states that require a small number of actions to accomplish the task, and expands the initial state distribution backwards in a dynamically-consistent manner once the policy optimization algorithm demonstrates sufficient performance.

Continuous Control reinforcement-learning +1

Safe Motion Planning in Unknown Environments: Optimality Benchmarks and Tractable Policies

no code implementations16 Apr 2018 Lucas Janson, Tommy Hu, Marco Pavone

This paper addresses the problem of planning a safe (i. e., collision-free) trajectory from an initial state to a goal region when the obstacle space is a-priori unknown and is incrementally revealed online, e. g., through line-of-sight perception.

Motion Planning

Generative Modeling of Multimodal Multi-Human Behavior

1 code implementation6 Mar 2018 Boris Ivanovic, Edward Schmerling, Karen Leung, Marco Pavone

This work presents a methodology for modeling and predicting human behavior in settings with N humans interacting in highly multimodal scenarios (i. e. where there are many possible highly-distinct futures).

Robotics Human-Computer Interaction

Risk-sensitive Inverse Reinforcement Learning via Semi- and Non-Parametric Methods

1 code implementation28 Nov 2017 Sumeet Singh, Jonathan Lacotte, Anirudha Majumdar, Marco Pavone

The literature on Inverse Reinforcement Learning (IRL) typically assumes that humans take actions in order to minimize the expected value of a cost function, i. e., that humans are risk neutral.

Decision Making reinforcement-learning +1

How Should a Robot Assess Risk? Towards an Axiomatic Theory of Risk in Robotics

no code implementations30 Oct 2017 Anirudha Majumdar, Marco Pavone

We discuss general representation theorems that precisely characterize the class of metrics that satisfy these axioms (referred to as distortion risk metrics), and provide instantiations that can be used in applications.

Decision Making

Multimodal Probabilistic Model-Based Planning for Human-Robot Interaction

1 code implementation25 Oct 2017 Edward Schmerling, Karen Leung, Wolf Vollprecht, Marco Pavone

This paper presents a method for constructing human-robot interaction policies in settings where multimodality, i. e., the possibility of multiple highly distinct futures, plays a critical role in decision making.

Decision Making

Data-Driven Model Predictive Control of Autonomous Mobility-on-Demand Systems

no code implementations20 Sep 2017 Ramon Iglesias, Federico Rossi, Kevin Wang, David Hallac, Jure Leskovec, Marco Pavone

The goal of this paper is to present an end-to-end, data-driven framework to control Autonomous Mobility-on-Demand systems (AMoD, i. e. fleets of self-driving vehicles).

Robotics Multiagent Systems Systems and Control Applications

Learning Sampling Distributions for Robot Motion Planning

2 code implementations16 Sep 2017 Brian Ichter, James Harrison, Marco Pavone

This paper proposes a methodology for non-uniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling.

Collision Avoidance Motion Planning

On the interaction between Autonomous Mobility-on-Demand systems and the power network: models and coordination algorithms

1 code implementation14 Sep 2017 Federico Rossi, Ramon Iglesias, Mahnoosh Alizadeh, Marco Pavone

We assess the performance of the mode by studying a hypothetical AMoD system in Dallas-Fort Worth and its impact on the Texas power network.

Systems and Control Multiagent Systems Robotics

Mixed Strategy for Constrained Stochastic Optimal Control

no code implementations6 Jul 2016 Masahiro Ono, Mahmoud El Chamie, Marco Pavone, Behcet Acikmese

We found that the same result holds for stochastic optimal control problems with continuous state and action spaces. Furthermore, we show the randomization of control input can result in reduced cost when the optimization problem is nonconvex, and the cost reduction is equal to the duality gap.


Risk-Constrained Reinforcement Learning with Percentile Risk Criteria

no code implementations5 Dec 2015 Yin-Lam Chow, Mohammad Ghavamzadeh, Lucas Janson, Marco Pavone

In many sequential decision-making problems one is interested in minimizing an expected cumulative cost while taking into account \emph{risk}, i. e., increased awareness of events of small probability and high consequences.

Decision Making Marketing +2

Two Phase $Q-$learning for Bidding-based Vehicle Sharing

no code implementations29 Sep 2015 Yin-Lam Chow, Jia Yuan Yu, Marco Pavone

We consider one-way vehicle sharing systems where customers can rent a car at one station and drop it off at another.

Decision Making Q-Learning +1

Risk-Sensitive and Robust Decision-Making: a CVaR Optimization Approach

no code implementations NeurIPS 2015 Yin-Lam Chow, Aviv Tamar, Shie Mannor, Marco Pavone

Our first contribution is to show that a CVaR objective, besides capturing risk sensitivity, has an alternative interpretation as expected cost under worst-case modeling errors, for a given error budget.

Decision Making

Monte Carlo Motion Planning for Robot Trajectory Optimization Under Uncertainty

1 code implementation30 Apr 2015 Lucas Janson, Edward Schmerling, Marco Pavone

MCMP applies this CP estimation procedure to motion planning by iteratively (i) computing an (approximately) optimal path for the deterministic version of the problem (here, using the FMT* algorithm), (ii) computing the CP of this path, and (iii) inflating or deflating the obstacles by a common factor depending on whether the CP is higher or lower than a target value.


Cannot find the paper you are looking for? You can Submit a new open access paper.