arXiv subjects

Bo Cheng

Publications and source records attributed to Bo Cheng.

At least 55 records · Page 3Linked to original sources

arXiv2023

Inverted Landing in a Small Aerial Robot via Deep Reinforcement Learning for Triggering and Control of Rotational Maneuvers

Inverted landing in a rapid and robust manner is a challenging feat for aerial robots, especially while depending entirely on onboard sensing and computation. In spite of this, this feat is routinely performed by biological fliers such as bats, flies, and bees. Our previous work has identified a direct causal connection between a series of onboard visual cues and kinematic actions that allow for reliable execution of this challenging aerobatic maneuver in small aerial robots. In this work, we first utilized Deep Reinforcement Learning and a physics-based simulation to obtain a general, optimal control policy for robust inverted landing starting from any arbitrary approach condition. This optimized control policy provides a computationally-efficient mapping from the system's observational space to its motor command action space, including both triggering and control of rotational maneuvers. This was done by training the system over a large range of approach flight velocities that varied with magnitude and direction. Next, we performed a sim-to-real transfer and experimental validation of the learned policy via domain randomization, by varying the robot's inertial parameters in the simulation. Through experimental trials, we identified several dominant factors which greatly improved landing robustness and the primary mechanisms that determined inverted landing success. We expect the learning framework developed in this study can be generalized to solve more challenging tasks, such as utilizing noisy onboard sensory data, landing on surfaces of various orientations, or landing on dynamically-moving surfaces.

cs.RO↗

arXiv2022

Generative Prompt Tuning for Relation Classification

Using prompts to explore the knowledge contained within pre-trained language models for downstream tasks has now become an active topic. Current prompt tuning methods mostly convert the downstream tasks to masked language modeling problems by adding cloze-style phrases and mapping all labels to verbalizations with fixed length, which has proven effective for tasks with simple label spaces. However, when applied to relation classification exhibiting complex label spaces, vanilla prompt tuning methods may struggle with label verbalizations with arbitrary lengths due to rigid prompt restrictions. Inspired by the text infilling task for pre-training generative models that can flexibly predict missing spans, we propose a novel generative prompt tuning method to reformulate relation classification as an infilling problem, which frees our approach from limitations of current prompt based approaches and thus fully exploits rich semantics of entity and relation types. In addition, we design entity-guided decoding and discriminative relation scoring to generate and align relations effectively and efficiently during inference. Extensive experiments under fully supervised settings and low-resource settings demonstrate the effectiveness of our approach.

cs.CL↗

arXiv2022

Analyzing Modality Robustness in Multimodal Sentiment Analysis

Building robust multimodal models are crucial for achieving reliable deployment in the wild. Despite its importance, less attention has been paid to identifying and improving the robustness of Multimodal Sentiment Analysis (MSA) models. In this work, we hope to address that by (i) Proposing simple diagnostic checks for modality robustness in a trained multimodal model. Using these checks, we find MSA models to be highly sensitive to a single modality, which creates issues in their robustness; (ii) We analyze well-known robust training strategies to alleviate the issues. Critically, we observe that robustness can be achieved without compromising on the original performance. We hope our extensive study-performed across five models and two benchmark datasets-and proposed procedures would make robustness an integral component in MSA research. Our diagnostic checks and robust training solutions are simple to implement and available at https://github. com/declare-lab/MSA-Robustness.

cs.CL↗

arXiv2022

Exploring Entity Interactions for Few-Shot Relation Learning (Student Abstract)

Few-shot relation learning refers to infer facts for relations with a limited number of observed triples. Existing metric-learning methods for this problem mostly neglect entity interactions within and between triples. In this paper, we explore this kind of fine-grained semantic meanings and propose our model TransAM. Specifically, we serialize reference entities and query entities into sequence and apply transformer structure with local-global attention to capture both intra- and inter-triple entity interactions. Experiments on two public benchmark datasets NELL-One and Wiki-One with 1-shot setting prove the effectiveness of TransAM.

cs.CL↗

arXiv2022

Adaptive dynamic programming for nonaffine nonlinear optimal control problem with state constraints

This paper presents a constrained adaptive dynamic programming (CADP) algorithm to solve general nonlinear nonaffine optimal control problems with known dynamics. Unlike previous ADP algorithms, it can directly deal with problems with state constraints. Firstly, a constrained generalized policy iteration (CGPI) framework is developed to handle state constraints by transforming the traditional policy improvement process into a constrained policy optimization problem. Next, we propose an actor-critic variant of CGPI, called CADP, in which both policy and value functions are approximated by multi-layer neural networks to directly map the system states to control inputs and value function, respectively. CADP linearizes the constrained optimization problem locally into a quadratically constrained linear programming problem, and then obtains the optimal update of the policy network by solving its dual problem. A trust region constraint is added to prevent excessive policy update, thus ensuring linearization accuracy. We determine the feasibility of the policy optimization problem by calculating the minimum trust region boundary and update the policy using two recovery rules when infeasible. The vehicle control problem in the path-tracking task is used to demonstrate the effectiveness of this proposed method.

eess.SY↗

arXiv2022

Recurrent Model Predictive Control: Learning an Explicit Recurrent Controller for Nonlinear Systems

This paper proposes an offline control algorithm, called Recurrent Model Predictive Control (RMPC), to solve large-scale nonlinear finite-horizon optimal control problems. It can be regarded as an explicit solver of traditional Model Predictive Control (MPC) algorithms, which can adaptively select appropriate model prediction horizon according to current computing resources, so as to improve the policy performance. Our algorithm employs a recurrent function to approximate the optimal policy, which maps the system states and reference values directly to the control inputs. The output of the learned policy network after N recurrent cycles corresponds to the nearly optimal solution of N-step MPC. A policy optimization objective is designed by decomposing the MPC cost function according to the Bellman's principle of optimality. The optimal recurrent policy can be obtained by directly minimizing the designed objective function, which is applicable for general nonlinear and non input-affine systems. Both simulation-based and real-robot path-tracking tasks are utilized to demonstrate the effectiveness of the proposed method.

eess.SY↗

arXiv2022

Fixed-Dimensional and Permutation Invariant State Representation of Autonomous Driving

In this paper, we propose a new state representation method, called encoding sum and concatenation (ESC), for the state representation of decision-making in autonomous driving. Unlike existing state representation methods, ESC is applicable to a variable number of surrounding vehicles and eliminates the need for manually pre-designed sorting rules, leading to higher representation ability and generality. The proposed ESC method introduces a representation neural network (NN) to encode each surrounding vehicle into an encoding vector, and then adds these vectors to obtain the representation vector of the set of surrounding vehicles. By concatenating the set representation with other variables, such as indicators of the ego vehicle and road, we realize the fixed-dimensional and permutation invariant state representation. This paper has further proved that the proposed ESC method can realize the injective representation if the output dimension of the representation NN is greater than the number of variables of all surrounding vehicles. This means that by taking the ESC representation as policy inputs, we can find the nearly optimal representation NN and policy NN by simultaneously optimizing them using gradient-based updating. Experiments demonstrate that compared with the fixed-permutation representation method, the proposed method improves the representation ability of the surrounding vehicles, and the corresponding approximation error is reduced by 62.2%.

cs.RO↗

arXiv2022

Optimal Inverted Landing in a Small Aerial Robot with Varied Approach Velocities and Landing Gear Designs

Inverted landing is a challenging feat to perform in aerial robots, especially without external positioning. However, it is routinely performed by biological fliers such as bees, flies, and bats. Our previous observations of landing behaviors in flies suggest an open-loop causal relationship between their putative visual cues and the kinematics of the aerial maneuvers executed. For example, the degree of rotational maneuver (the amount of body inversion prior to touchdown) and the amount of leg-assisted body swing both depend on the flies' initial body states while approaching the ceiling. In this work, inspired by the inverted landing behavior of flies, we used a physics-based simulation with experimental validation to systematically investigate how optimized inverted landing maneuvers depend on the initial approach velocities with varied magnitude and direction. This was done by analyzing the putative visual cues (that can be derived from onboard measurements) during optimal maneuvering trajectories. We identified a three-dimensional policy region, from which a mapping to a global inverted landing policy can be developed without the use of external positioning data. Through simulation, we also investigated the effects of an array of landing gear designs on the optimized landing performance and identified their advantages and disadvantages. The above results have been partially validated using limited experimental testing and will continue to inform and guide our future experiments, for example by applying the calculated global policy.

cs.RO↗

arXiv2021

Development of a GPU-accelerated Monte Carlo dose calculation module for nuclear medicine, ARCHER-NM: Demonstration for a PET/CT imaging procedure

This paper describes the development and validation of a Monte Carlo (MC) dose computing module dedicated to organ dose calculations of patients undergoing nuclear medicine (NM) internal radiation exposures involving 18F-FDG PET/CT examination. This new module extends the more-than-10-years-long ARCHER project that developed a GPU-accelerated MC dose engine by adding dedicated NM source-definition features. To validate the code, we compared dose distributions from the 0.511-MeV point photon source calculated for a water phantom as well as a patient PET/CT phantom against a well-tested MC code, GATE. The water-phantom results show excellent agreement, suggesting that the radiation physics module in the new NM code is adequate. To demonstrate the clinical utility and advantage of ARCHER-NM, one set of PET/CT data for an adult male NM patient is calculated using the new code. Radiosensitive organs in the CT dataset are segmented using a CNN-based tool called DeepViewer. The PET image intensity maps are converted to radioactivity distributions to allow for MC radiation transport dose calculations at the voxel level. The dose rate maps and corresponding statistical uncertainties were calculated for the duration of PET image acquisition. The dose rate results of the 18F-FDG PET imaging patient show that ARCHER-NM's results agree very well with those of the GATE within 0.58% to 4.11%. Most impressively, ARCHER-NM obtains such results in less than 0.5 minutes while it takes GATE as much as 376 minutes. This is the first study presenting GPU-accelerated patient-specific MC internal radiation dose rate calculations for clinically realistic 18F-FDG PET/CT imaging cases involving auto-segmentation of whole-body PET/CT images. This study suggests that modern computing tools -- ARCHER-NM and DeepViewer -- are accurate and fast enough for routine internal dosimetry in NM clinics.

physics.med-ph↗

arXiv2021

Effects of Design and Hydrodynamic Parameters on Optimized Swimming for Simulated, Fish-inspired Robots

In this work we developed a mathematical model and a simulation platform for a fish-inspired robotic template, namely Magnetic, Modular, Undulatory Robotics ($μ$Bots). Through this platform, we systematically explored the effects of design and fluid parameters on the swimming performance via reinforcement learning. The mathematical model was composed of two interacting subsystems, the robot dynamics and the hydrodynamics, and the hydrodynamic model consisted of reactive components (added-mass and pressure forces) and resistive components (drag and friction forces), which were then nondimensionalized for deriving key "control parameters" of robot-fluid interaction. The $μ$Bot was actuated via magnetic actuators controlled with harmonic voltage signals, which were optimized via EM-based Policy Hyper Parameter Exploration (EPHE) to maximize swimming speed. By varying the control parameters, total 36 cases with different robot template variations (Number of Actuation (NoA) and stiffness) and hydrodynamic parameters were simulated and optimized via EPHE. Results showed that wavelength of optimized gaits (i.e., traveling wave along body) was independent of template variations and hydrodynamic parameters. Higher NoA yielded higher speed but lower speed per body length however with diminishing gain and lower speed per body length. Body and caudal-fin gait dynamics were dominated by the interaction among fluid added-mass, spring, and actuation torque, with negligible contribution from fluid resistive drag. In contrast, thrust generation was dominated by pressure force acting on caudal fin, as steady swimming resulted from a balance between resistive force and pressure force, with minor contributions from added-mass and body drag forces. Therefore, added-mass force only indirectly affected the thrust generation and swimming speed via the caudal fin dynamics.

cs.RO↗

arXiv2021

Exploring Task Difficulty for Few-Shot Relation Extraction

Few-shot relation extraction (FSRE) focuses on recognizing novel relations by learning with merely a handful of annotated instances. Meta-learning has been widely adopted for such a task, which trains on randomly generated few-shot tasks to learn generic data representations. Despite impressive results achieved, existing models still perform suboptimally when handling hard FSRE tasks, where the relations are fine-grained and similar to each other. We argue this is largely because existing models do not distinguish hard tasks from easy ones in the learning process. In this paper, we introduce a novel approach based on contrastive learning that learns better representations by exploiting relation label information. We further design a method that allows the model to adaptively learn how to focus on hard tasks. Experiments on two standard datasets demonstrate the effectiveness of our method.

cs.CL↗

arXiv2021

FCM: A Fine-grained Comparison Model for Multi-turn Dialogue Reasoning

Despite the success of neural dialogue systems in achieving high performance on the leader-board, they cannot meet users' requirements in practice, due to their poor reasoning skills. The underlying reason is that most neural dialogue models only capture the syntactic and semantic information, but fail to model the logical consistency between the dialogue history and the generated response. Recently, a new multi-turn dialogue reasoning task has been proposed, to facilitate dialogue reasoning research. However, this task is challenging, because there are only slight differences between the illogical response and the dialogue history. How to effectively solve this challenge is still worth exploring. This paper proposes a Fine-grained Comparison Model (FCM) to tackle this problem. Inspired by human's behavior in reading comprehension, a comparison mechanism is proposed to focus on the fine-grained differences in the representation of each response candidate. Specifically, each candidate representation is compared with the whole history to obtain a history consistency representation. Furthermore, the consistency signals between each candidate and the speaker's own history are considered to drive a model to prefer a candidate that is logically consistent with the speaker's history logic. Finally, the above consistency representations are employed to output a ranking list of the candidate responses for multi-turn dialogue reasoning. Experimental results on two public dialogue datasets show that our method obtains higher ranking scores than the baseline models.

cs.CL↗

arXiv2021

Encoding Distributional Soft Actor-Critic for Autonomous Driving in Multi-lane Scenarios

In this paper, we propose a new reinforcement learning (RL) algorithm, called encoding distributional soft actor-critic (E-DSAC), for decision-making in autonomous driving. Unlike existing RL-based decision-making methods, E-DSAC is suitable for situations where the number of surrounding vehicles is variable and eliminates the requirement for manually pre-designed sorting rules, resulting in higher policy performance and generality. We first develop an encoding distributional policy iteration (DPI) framework by embedding a permutation invariant module, which employs a feature neural network (NN) to encode the indicators of each vehicle, in the distributional RL framework. The proposed DPI framework is proved to exhibit important properties in terms of convergence and global optimality. Next, based on the developed encoding DPI framework, we propose the E-DSAC algorithm by adding the gradient-based update rule of the feature NN to the policy evaluation process of the DSAC algorithm. Then, the multi-lane driving task and the corresponding reward function are designed to verify the effectiveness of the proposed algorithm. Results show that the policy learned by E-DSAC can realize efficient, smooth, and relatively safe autonomous driving in the designed scenario. And the final policy performance learned by E-DSAC is about three times that of DSAC. Furthermore, its effectiveness has also been verified in real vehicle experiments.

cs.RO↗

arXiv2021

State-space aerodynamic model reveals high force control authority and predictability in flapping flight

Flying animals resort to fast, large-degree-of-freedom motion of flapping wings, a key feature that distinguishes them from rotary or fixed-winged robotic fliers with limited motion of aerodynamic surfaces. However, flapping-wing aerodynamics are characterised by highly unsteady and three-dimensional flows difficult to model or control, and accurate aerodynamic force predictions often rely on expensive computational or experimental methods. Here, we developed a computationally efficient and data-driven state-space model to dynamically map wing kinematics to aerodynamic forces/moments. This model was trained and tested with a total of 548 different flapping-wing motions and surpassed the accuracy and generality of the existing quasi-steady models. This model used 12 states to capture the unsteady and nonlinear fluid effects pertinent to force generation without explicit information of fluid flows. We also provided a comprehensive assessment of the control authority of key wing kinematic variables and found that instantaneous aerodynamic forces/moments were largely predictable by the wing motion history within a half-stroke cycle. Furthermore, the angle of attack, normal acceleration, and pitching motion had the strongest effects on the aerodynamic force/moment generation. Our results show that flapping flight inherently offers high force control authority and predictability, which can be key to developing agile and stable aerial fliers.

physics.flu-dyn↗

arXiv2021

Distributional Soft Actor-Critic: Off-Policy Reinforcement Learning for Addressing Value Estimation Errors

In reinforcement learning (RL), function approximation errors are known to easily lead to the Q-value overestimations, thus greatly reducing policy performance. This paper presents a distributional soft actor-critic (DSAC) algorithm, which is an off-policy RL method for continuous control setting, to improve the policy performance by mitigating Q-value overestimations. We first discover in theory that learning a distribution function of state-action returns can effectively mitigate Q-value overestimations because it is capable of adaptively adjusting the update stepsize of the Q-value function. Then, a distributional soft policy iteration (DSPI) framework is developed by embedding the return distribution function into maximum entropy RL. Finally, we present a deep off-policy actor-critic variant of DSPI, called DSAC, which directly learns a continuous return distribution by keeping the variance of the state-action returns within a reasonable range to address exploding and vanishing gradient problems. We evaluate DSAC on the suite of MuJoCo continuous control tasks, achieving the state-of-the-art performance.

cs.LG↗

arXiv2021

DPN-SENet:A self-attention mechanism neural network for detection and diagnosis of COVID-19 from chest x-ray images

Background and Objective: The new type of coronavirus is also called COVID-19. It began to spread at the end of 2019 and has now spread across the world. Until October 2020, It has infected around 37 million people and claimed about 1 million lives. We propose a deep learning model that can help radiologists and clinicians use chest X-rays to diagnose COVID-19 cases and show the diagnostic features of pneumonia. Methods: The approach in this study is: 1) we propose a data enhancement method to increase the diversity of the data set, thereby improving the generalization performance of the model. 2) Our deep convolution neural network model DPN-SE adds a self-attention mechanism to the DPN network. The addition of a self-attention mechanism has greatly improved the performance of the network. 3) Use the Lime interpretable library to mark the feature regions on the X-ray medical image that helps doctors more quickly diagnose COVID-19 in people. Results: Under the same network model, the data with and without data enhancement is put into the model for training respectively. At last, comparing two experimental results: among the 10 network models with different structures, 7 network models have improved their effects after using data enhancement, with an average improvement of 1% in recognition accuracy. We propose that the accuracy and recall rates of the DPN-SE network are 93% and 98% of cases (COVID vs. pneumonia bacteria vs. viral pneumonia vs. normal). Compared with the original DPN, the respective accuracy is improved by 2%. Conclusion: The data augmentation method we used has achieved effective results on a small amount of data set, showing that a reasonable data augmentation method can improve the recognition accuracy without changing the sample size and model structure. Overall, the proposed method and model can effectively become a very useful tool for clinical radiologists.

eess.IV↗

arXiv2021

Direct and indirect reinforcement learning

Reinforcement learning (RL) algorithms have been successfully applied to a range of challenging sequential decision making and control tasks. In this paper, we classify RL into direct and indirect RL according to how they seek the optimal policy of the Markov decision process problem. The former solves the optimal policy by directly maximizing an objective function using gradient descent methods, in which the objective function is usually the expectation of accumulative future rewards. The latter indirectly finds the optimal policy by solving the Bellman equation, which is the sufficient and necessary condition from Bellman's principle of optimality. We study policy gradient forms of direct and indirect RL and show that both of them can derive the actor-critic architecture and can be unified into a policy gradient with the approximate value function and the stationary state distribution, revealing the equivalence of direct and indirect RL. We employ a Gridworld task to verify the influence of different forms of policy gradient, suggesting their differences and relationships experimentally. Finally, we classify current mainstream RL algorithms using the direct and indirect taxonomy, together with other ones including value-based and policy-based, model-based and model-free.

cs.LG↗

arXiv2021

Integrated Decision and Control: Towards Interpretable and Computationally Efficient Driving Intelligence

Decision and control are core functionalities of high-level automated vehicles. Current mainstream methods, such as functionality decomposition and end-to-end reinforcement learning (RL), either suffer high time complexity or poor interpretability and adaptability on real-world autonomous driving tasks. In this paper, we present an interpretable and computationally efficient framework called integrated decision and control (IDC) for automated vehicles, which decomposes the driving task into static path planning and dynamic optimal tracking that are structured hierarchically. First, the static path planning generates several candidate paths only considering static traffic elements. Then, the dynamic optimal tracking is designed to track the optimal path while considering the dynamic obstacles. To that end, we formulate a constrained optimal control problem (OCP) for each candidate path, optimize them separately and follow the one with the best tracking performance. To unload the heavy online computation, we propose a model-based reinforcement learning (RL) algorithm that can be served as an approximate constrained OCP solver. Specifically, the OCPs for all paths are considered together to construct a single complete RL problem and then solved offline in the form of value and policy networks, for real-time online path selecting and tracking respectively. We verify our framework in both simulations and the real world. Results show that compared with baseline methods IDC has an order of magnitude higher online computing efficiency, as well as better driving performance including traffic efficiency and safety. In addition, it yields great interpretability and adaptability among different driving tasks. The effectiveness of the proposed method is also demonstrated in real road tests with complicated traffic conditions.

cs.LG↗