arXiv subjects

Sertac Karaman

Publications and source records attributed to Sertac Karaman.

At least 55 records · Page 3Linked to original sources

arXiv2021

VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles

Simulation has the potential to transform the development of robust algorithms for mobile agents deployed in safety-critical scenarios. However, the poor photorealism and lack of diverse sensor modalities of existing simulation engines remain key hurdles towards realizing this potential. Here, we present VISTA, an open source, data-driven simulator that integrates multiple types of sensors for autonomous vehicles. Using high fidelity, real-world datasets, VISTA represents and simulates RGB cameras, 3D LiDAR, and event-based cameras, enabling the rapid generation of novel viewpoints in simulation and thereby enriching the data available for policy learning with corner cases that are difficult to capture in the physical world. Using VISTA, we demonstrate the ability to train and test perception-to-control policies across each of the sensor types and showcase the power of this approach via deployment on a full scale autonomous vehicle. The policies learned in VISTA exhibit sim-to-real transfer without modification and greater robustness than those trained exclusively on real-world data.

cs.RO↗

arXiv2021

Learning Interactive Driving Policies via Data-driven Simulation

Data-driven simulators promise high data-efficiency for driving policy learning. When used for modelling interactions, this data-efficiency becomes a bottleneck: Small underlying datasets often lack interesting and challenging edge cases for learning interactive driving. We address this challenge by proposing a simulation method that uses in-painted ado vehicles for learning robust driving policies. Thus, our approach can be used to learn policies that involve multi-agent interactions and allows for training via state-of-the-art policy learning methods. We evaluate the approach for learning standard interaction scenarios in driving. In extensive experiments, our work demonstrates that the resulting policies can be directly transferred to a full-scale autonomous vehicle without making use of any traditional sim-to-real transfer techniques such as domain randomization.

cs.RO↗

arXiv2021

Searching for Efficient Multi-Stage Vision Transformers

Vision Transformer (ViT) demonstrates that Transformer for natural language processing can be applied to computer vision tasks and result in comparable performance to convolutional neural networks (CNN), which have been studied and adopted in computer vision for years. This naturally raises the question of how the performance of ViT can be advanced with design techniques of CNN. To this end, we propose to incorporate two techniques and present ViT-ResNAS, an efficient multi-stage ViT architecture designed with neural architecture search (NAS). First, we propose residual spatial reduction to decrease sequence lengths for deeper layers and utilize a multi-stage architecture. When reducing lengths, we add skip connections to improve performance and stabilize training deeper networks. Second, we propose weight-sharing NAS with multi-architectural sampling. We enlarge a network and utilize its sub-networks to define a search space. A super-network covering all sub-networks is then trained for fast evaluation of their performance. To efficiently train the super-network, we propose to sample and train multiple sub-networks with one forward-backward pass. After that, evolutionary search is performed to discover high-performance network architectures. Experiments on ImageNet demonstrate that ViT-ResNAS achieves better accuracy-MACs and accuracy-throughput trade-offs than the original DeiT and other strong baselines of ViT. Code is available at https://github.com/yilunliao/vit-search.

cs.CV↗

arXiv2021

FlightGoggles: A Modular Framework for Photorealistic Camera, Exteroceptive Sensor, and Dynamics Simulation

FlightGoggles is a photorealistic sensor simulator for perception-driven robotic vehicles. The key contributions of FlightGoggles are twofold. First, FlightGoggles provides photorealistic exteroceptive sensor simulation using graphics assets generated with photogrammetry. Second, it provides the ability to combine (i) synthetic exteroceptive measurements generated in silico in real time and (ii) vehicle dynamics and proprioceptive measurements generated in motio by vehicle(s) in a motion-capture facility. FlightGoggles is capable of simulating a virtual-reality environment around autonomous vehicle(s). While a vehicle is in flight in the FlightGoggles virtual reality environment, exteroceptive sensors are rendered synthetically in real time while all complex extrinsic dynamics are generated organically through the natural interactions of the vehicle. The FlightGoggles framework allows for researchers to accelerate development by circumventing the need to estimate complex and hard-to-model interactions such as aerodynamics, motor mechanics, battery electrochemistry, and behavior of other agents. The ability to perform vehicle-in-the-loop experiments with photorealistic exteroceptive sensor simulation facilitates novel research directions involving, e.g., fast and agile autonomous flight in obstacle-rich environments, safe human interaction, and flexible sensor selection. FlightGoggles has been utilized as the main test for selecting nine teams that will advance in the AlphaPilot autonomous drone racing challenge. We survey approaches and results from the top AlphaPilot teams, which may be of independent interest.

cs.RO↗

arXiv2021

Efficient and Robust LiDAR-Based End-to-End Navigation

Deep learning has been used to demonstrate end-to-end neural network learning for autonomous vehicle control from raw sensory input. While LiDAR sensors provide reliably accurate information, existing end-to-end driving solutions are mainly based on cameras since processing 3D data requires a large memory footprint and computation cost. On the other hand, increasing the robustness of these systems is also critical; however, even estimating the model's uncertainty is very challenging due to the cost of sampling-based methods. In this paper, we present an efficient and robust LiDAR-based end-to-end navigation framework. We first introduce Fast-LiDARNet that is based on sparse convolution kernel optimization and hardware-aware model design. We then propose Hybrid Evidential Fusion that directly estimates the uncertainty of the prediction from only a single forward pass and then fuses the control predictions intelligently. We evaluate our system on a full-scale vehicle and demonstrate lane-stable as well as navigation capabilities. In the presence of out-of-distribution events (e.g., sensor failures), our system significantly improves robustness and reduces the number of takeovers in the real world.

cs.RO↗

arXiv2021

Stochastic Dynamic Games in Belief Space

Information gathering while interacting with other agents under sensing and motion uncertainty is critical in domains such as driving, service robots, racing, or surveillance. The interests of agents may be at odds with others, resulting in a stochastic non-cooperative dynamic game. Agents must predict others' future actions without communication, incorporate their actions into these predictions, account for uncertainty and noise in information gathering, and consider what information their actions reveal. Our solution uses local iterative dynamic programming in Gaussian belief space to solve a game-theoretic continuous POMDP. Solving a quadratic game in the backward pass of a game-theoretic belief-space variant of iLQG achieves a runtime polynomial in the number of agents and linear in the planning horizon. Our algorithm yields linear feedback policies for our robot, and predicted feedback policies for other agents. We present three applications: active surveillance, guiding eyes for a blind agent, and autonomous racing. Agents with game-theoretic belief-space planning win 44% more races than without game theory and 34% more than without belief-space planning.

cs.RO↗

arXiv2021

Feedback from Pixels: Output Regulation via Learning-Based Scene View Synthesis

We propose a novel controller synthesis involving feedback from pixels, whereby the measurement is a high dimensional signal representing a pixelated image with Red-Green-Blue (RGB) values. The approach neither requires feature extraction, nor object detection, nor visual correspondence. The control policy does not involve the estimation of states or similar latent representations. Instead, tracking is achieved directly in image space, with a model of the reference signal embedded as required by the internal model principle. The reference signal is generated by a neural network with learning-based scene view synthesis capabilities. Our approach does not require an end-to-end learning of a pixel-to-action control policy. The approach is applied to a motion control problem, namely the longitudinal dynamics of a car-following problem. We show how this approach lend itself to a tractable stability analysis with associated bounds critical to establishing trustworthiness and interpretability of the closed-loop dynamics.

eess.SY↗

arXiv2021

Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space

Learning competitive behaviors in multi-agent settings such as racing requires long-term reasoning about potential adversarial interactions. This paper presents Deep Latent Competition (DLC), a novel reinforcement learning algorithm that learns competitive visual control policies through self-play in imagination. The DLC agent imagines multi-agent interaction sequences in the compact latent space of a learned world model that combines a joint transition function with opponent viewpoint prediction. Imagined self-play reduces costly sample generation in the real world, while the latent representation enables planning to scale gracefully with observation dimensionality. We demonstrate the effectiveness of our algorithm in learning competitive behaviors on a novel multi-agent racing benchmark that requires planning from image observations. Code and videos available at https://sites.google.com/view/deep-latent-competition.

cs.LG↗

arXiv2021

Recovering orthogonal tensors under arbitrarily strong, but locally correlated, noise

We consider the problem of recovering an orthogonally decomposable tensor with a subset of elements distorted by noise with arbitrarily large magnitude. We focus on the particular case where each mode in the decomposition is corrupted by noise vectors with components that are correlated locally, i.e., with nearby components. We show that this deterministic tensor completion problem has the unusual property that it can be solved in polynomial time if the rank of the tensor is sufficiently large. This is the polar opposite of the low-rank assumptions of typical low-rank tensor and matrix completion settings. We show that our problem can be solved through a system of coupled Sylvester-like equations and show how to accelerate their solution by an alternating solver. This enables recovery even with a substantial number of missing entries, for instance for $n$-dimensional tensors of rank $n$ with up to $40\%$ missing entries.

math.NA↗

arXiv2020

Low Cost, Educational Internal Combustion Engine Electronic Control Unit Hardware-in-the-Loop Test Systems

Different hardware platforms and their associated real time operating systems that can be used in an educational laboratory for illustrating engine electronic control unit hardware in the loop testing are presented and compared in this paper. A Matlab graphical user interface prepared for generating synthetic crank and camshaft angular position sensor signals to be fed to the engine electronic control unit during hardware-in-the-loop testing is introduced. This graphical user interface is used to generate faulty sensor signals to check the response of the engine electronic control unit during hardware-in-the-loop simulation. Examples of faulty signals that can be generated with the graphical user interface are illustrated.

eess.SY↗

arXiv2020

Multi-resolution Low-rank Tensor Formats

We describe a simple, black-box compression format for tensors with a multiscale structure. By representing the tensor as a sum of compressed tensors defined on increasingly coarse grids, we capture low-rank structures on each grid-scale, and we show how this leads to an increase in compression for a fixed accuracy. We devise an alternating algorithm to represent a given tensor in the multiresolution format and prove local convergence guarantees. In two dimensions, we provide examples that show that this approach can beat the Eckart-Young theorem, and for dimensions higher than two, we achieve higher compression than the tensor-train format on six real-world datasets. We also provide results on the closedness and stability of the tensor format and discuss how to perform common linear algebra operations on the level of the compressed tensors.

math.NA↗

arXiv2020

Deep Context Maps: Agent Trajectory Prediction using Location-specific Latent Maps

In this paper, we propose a novel approach for agent motion prediction in cluttered environments. One of the main challenges in predicting agent motion is accounting for location and context-specific information. Our main contribution is the concept of learning context maps to improve the prediction task. Context maps are a set of location-specific latent maps that are trained alongside the predictor. Thus, the proposed maps are capable of capturing location context beyond visual context cues (e.g. usual average speeds and typical trajectories) or predefined map primitives (such as lanes and stop lines). We pose context map learning as a multi-task training problem and describe our map model and its incorporation into a state-of-the-art trajectory predictor. In extensive experiments, it is shown that use of learned maps can significantly improve predictor accuracy. Furthermore, the performance can be additionally boosted by providing partial knowledge of map semantics.

cs.RO↗

arXiv2020

Accurate Tracking of Aggressive Quadrotor Trajectories using Incremental Nonlinear Dynamic Inversion and Differential Flatness

Autonomous unmanned aerial vehicles (UAVs) that can execute aggressive (i.e., high-speed and high-acceleration) maneuvers have attracted significant attention in the past few years. This paper focuses on accurate tracking of aggressive quadcopter trajectories. We propose a novel control law for tracking of position and yaw angle and their derivatives of up to fourth order, specifically, velocity, acceleration, jerk, and snap along with yaw rate and yaw acceleration. Jerk and snap are tracked using feedforward inputs for angular rate and angular acceleration based on the differential flatness of the quadcopter dynamics. Snap tracking requires direct control of body torque, which we achieve using closed-loop motor speed control based on measurements from optical encoders attached to the motors. The controller utilizes incremental nonlinear dynamic inversion (INDI) for robust tracking of linear and angular accelerations despite external disturbances, such as aerodynamic drag forces. Hence, prior modeling of aerodynamic effects is not required. We rigorously analyze the proposed control law through response analysis, and we demonstrate it in experiments. The controller enables a quadcopter UAV to track complex 3D trajectories, reaching speeds up to 12.9 m/s and accelerations up to 2.1g, while keeping the root-mean-square tracking error down to 6.6 cm, in a flight volume that is roughly 18 m by 7 m and 3 m tall. We also demonstrate the robustness of the controller by attaching a drag plate to the UAV in flight tests and by pulling on the UAV with a rope during hover.

cs.RO↗

arXiv2020

Multi-Fidelity Black-Box Optimization for Time-Optimal Quadrotor Maneuvers

We consider the problem of generating a time-optimal quadrotor trajectory that attains a set of prescribed waypoints. This problem is challenging since the optimal trajectory is located on the boundary of the set of dynamically feasible trajectories. This boundary is hard to model as it involves limitations of the entire system, including hardware and software, in agile high-speed flight. In this work, we propose a multi-fidelity Bayesian optimization framework that models the feasibility constraints based on analytical approximation, numerical simulation, and real-world flight experiments. By combining evaluations at different fidelities, trajectory time is optimized while keeping the number of required costly flight experiments to a minimum. The algorithm is thoroughly evaluated in both simulation and real-world flight experiments at speeds up to 11 m/s. Resulting trajectories were found to be significantly faster than those obtained through minimum-snap trajectory planning.

cs.RO↗

arXiv2020

Perception-aware time optimal path parameterization for quadrotors

The increasing popularity of quadrotors has given rise to a class of predominantly vision-driven vehicles. This paper addresses the problem of perception-aware time optimal path parametrization for quadrotors. Although many different choices of perceptual modalities are available, the low weight and power budgets of quadrotor systems makes a camera ideal for on-board navigation and estimation algorithms. However, this does come with a set of challenges. The limited field of view of the camera can restrict the visibility of salient regions in the environment, which dictates the necessity to consider perception and planning jointly. The main contribution of this paper is an efficient time optimal path parametrization algorithm for quadrotors with limited field of view constraints. We show in a simulation study that a state-of-the-art controller can track planned trajectories, and we validate the proposed algorithm on a quadrotor platform in experiments.

cs.RO↗

arXiv2020

On Algorithms for and Computing with the Tensor Ring Decomposition

Tensor decompositions such as the canonical format and the tensor train format have been widely utilized to reduce storage costs and operational complexities for high-dimensional data, achieving linear scaling with the input dimension instead of exponential scaling. In this paper, we investigate even lower storage-cost representations in the tensor ring format, which is an extension of the tensor train format with variable end-ranks. Firstly, we introduce two algorithms for converting a tensor in full format to tensor ring format with low storage cost. Secondly, we detail a rounding operation for tensor rings and show how this requires new definitions of common linear algebra operations in the format to obtain storage-cost savings. Lastly, we introduce algorithms for transforming the graph structure of graph-based tensor formats, with orders of magnitude lower complexity than existing literature. The efficiency of all algorithms is demonstrated on a number of numerical examples, and in certain cases, we demonstrate significantly higher compression ratios when compared to previous approaches to using the tensor ring format.

math.NA↗

arXiv2019

Optimizing Information Freshness in Wireless Networks under General Interference Constraints

Age of information (AoI) is a recently proposed metric for measuring information freshness. AoI measures the time that elapsed since the last received update was generated. We consider the problem of minimizing average and peak AoI in a wireless networks, consisting of a set of source-destination links, under general interference constraints. When fresh information is always available for transmission, we show that a stationary scheduling policy is peak age optimal. We also prove that this policy achieves average age that is within a factor of two of the optimal average age. In the case where fresh information is not always available, and packet/information generation rate has to be controlled along with scheduling links for transmission, we prove an important separation principle: the optimal scheduling policy can be designed assuming fresh information, and independently, the packet generation rate control can be done by ignoring interference. Peak and average AoI for discrete time G/Ber/1 queue is analyzed for the first time, which may be of independent interest.

cs.IT↗

arXiv2019

A Theory of Uncertainty Variables for State Estimation and Inference

We develop a new framework of uncertainty variables to model uncertainty. An uncertainty variable is characterized by an uncertainty set, in which its realization is bound to lie, while the conditional uncertainty is characterized by a set map, from a given realization of a variable to a set of possible realizations of another variable. We prove Bayes' law and the law of total probability equivalents for uncertainty variables. We define a notion of independence, conditional independence, and pairwise independence for a collection of uncertainty variables, and show that this new notion of independence preserves the properties of independence defined over random variables. We then develop a graphical model, namely Bayesian uncertainty network, a Bayesian network equivalent defined over a collection of uncertainty variables, and show that all the natural conditional independence properties, expected out of a Bayesian network, hold for the Bayesian uncertainty network. We also define the notion of point estimate, and show its relation with the maximum a posteriori estimate. Probability theory starts with a distribution function (equivalently a probability measure) as a primitive and builds all other useful concepts, such as law of total probability, Bayes' law, independence, graphical models, point estimate, on it. Our work shows that it is perfectly possible to start with a set, instead of a distribution function, and retain all the useful ideas needed for state estimation and inference.

stat.ML↗