Despite its importance in both industrial and service robotics, mobile manipulation remains a significant challenge as it requires a seamless integration of end-effector trajectory generation with navigation skills as well as reasoning over long-horizons. Existing methods struggle to control the large configuration space, and to navigate dynamic and unknown environments. In previous work, we proposed to decompose mobile manipulation tasks into a simplified motion generator for the end-effector in task space and a trained reinforcement learning agent for the mobile base to account for kinematic feasibility of the motion. In this work, we introduce Neural Navigation for Mobile Manipulation (N$^2$M$^2$) which extends this decomposition to complex obstacle environments and enables it to tackle a broad range of tasks in real world settings. The resulting approach can perform unseen, long-horizon tasks in unexplored environments while instantly reacting to dynamic obstacles and environmental changes. At the same time, it provides a simple way to define new mobile manipulation tasks. We demonstrate the capabilities of our proposed approach in extensive simulation and real-world experiments on multiple kinematically diverse mobile manipulators. Code and videos are publicly available at http://mobile-rl.cs.uni-freiburg.de.

Solving control tasks in complex environments automatically through learning offers great potential. While contemporary techniques from deep reinforcement learning (DRL) provide effective solutions, their decision-making is not transparent. We aim to provide insights into the decisions faced by the agent by learning an automaton model of environmental behavior under the control of an agent. However, for most control problems, automata learning is not scalable enough to learn a useful model. In this work, we raise the capabilities of automata learning such that it is possible to learn models for environments that have complex and continuous dynamics. The core of the scalability of our method lies in the computation of an abstract state-space representation, by applying dimensionality reduction and clustering on the observed environmental state space. The stochastic transitions are learned via passive automata learning from observed interactions of the agent and the environment. In an iterative model-based RL process, we sample additional trajectories to learn an accurate environment model in the form of a discrete-state Markov decision process (MDP). We apply our automata learning framework on popular RL benchmarking environments in the OpenAI Gym, including LunarLander, CartPole, Mountain Car, and Acrobot. Our results show that the learned models are so precise that they enable the computation of policies solving the respective control tasks. Yet the models are more concise and more general than neural-network-based policies and by using MDPs we benefit from a wealth of tools available for analyzing them. When solving the task of LunarLander, the learned model even achieved similar or higher rewards than deep RL policies learned with stable-baselines3.

Sparse knowledge graph (KG) scenarios pose a challenge for previous Knowledge Graph Completion (KGC) methods, that is, the completion performance decreases rapidly with the increase of graph sparsity. This problem is also exacerbated because of the widespread existence of sparse KGs in practical applications. To alleviate this challenge, we present a novel framework, LR-GCN, that is able to automatically capture valuable long-range dependency among entities to supplement insufficient structure features and distill logical reasoning knowledge for sparse KGC. The proposed approach comprises two main components: a GNN-based predictor and a reasoning path distiller. The reasoning path distiller explores high-order graph structures such as reasoning paths and encodes them as rich-semantic edges, explicitly compositing long-range dependencies into the predictor. This step also plays an essential role in densifying KGs, effectively alleviating the sparse issue. Furthermore, the path distiller further distills logical reasoning knowledge from these mined reasoning paths into the predictor. These two components are jointly optimized using a well-designed variational EM algorithm. Extensive experiments and analyses on four sparse benchmarks demonstrate the effectiveness of our proposed method.

Perception algorithms that provide estimates of their uncertainty are crucial to the development of autonomous robots that can operate in challenging and uncontrolled environments. Such perception algorithms provide the means for having risk-aware robots that reason about the probability of successfully completing a task when planning. There exist perception algorithms that come with models of their uncertainty; however, these models are often developed with assumptions, such as perfect data associations, that do not hold in the real world. Hence the resultant estimated uncertainty is a weak lower bound. To tackle this problem we present introspective perception - a novel approach for predicting accurate estimates of the uncertainty of perception algorithms deployed on mobile robots. By exploiting sensing redundancy and consistency constraints naturally present in the data collected by a mobile robot, introspective perception learns an empirical model of the error distribution of perception algorithms in the deployment environment and in an autonomously supervised manner. In this paper, we present the general theory of introspective perception and demonstrate successful implementations for two different perception tasks. We provide empirical results on challenging real-robot data for introspective stereo depth estimation and introspective visual simultaneous localization and mapping and show that they learn to predict their uncertainty with high accuracy and leverage this information to significantly reduce state estimation errors for an autonomous mobile robot.

Learning from humans allows non-experts to program robots with ease, lowering the resources required to build complex robotic solutions. Nevertheless, such data-driven approaches often lack the ability to provide guarantees regarding their learned behaviors, which is critical for avoiding failures and/or accidents. In this work, we focus on reaching/point-to-point motions, where robots must always reach their goal, independently of their initial state. This can be achieved by modeling motions as dynamical systems and ensuring that they are globally asymptotically stable. Hence, we introduce a novel Contrastive Learning loss for training Deep Neural Networks (DNN) that, when used together with an Imitation Learning loss, enforces the aforementioned stability in the learned motions. Differently from previous work, our method does not restrict the structure of its function approximator, enabling its use with arbitrary DNNs and allowing it to learn complex motions with high accuracy. We validate it using datasets and a real robot. In the former case, motions are 2 and 4 dimensional, modeled as first- and second-order dynamical systems. In the latter, motions are 3, 4, and 6 dimensional, of first and second order, and are used to control a 7DoF robot manipulator in its end effector space and joint space. More details regarding the real-world experiments are presented in: \url{https://youtu.be/OM-2edHBRfc}.

This paper addresses a novel multi-agent deep reinforcement learning (MADRL)-based positioning algorithm for multiple unmanned aerial vehicles (UAVs) collaboration (i.e., UAVs work as mobile base stations). The primary objective of the proposed algorithm is to establish dependable mobile access networks for cellular vehicle-to-everything (C-V2X) communication, thereby facilitating the realization of high-quality intelligent transportation systems (ITS). The reliable mobile access services can be achieved in following two ways, i.e., i) energy-efficient UAV operation and ii) reliable wireless communication services. For energy-efficient UAV operation, the reward of our proposed MADRL algorithm contains the features for UAV energy consumption models in order to realize efficient operations. Furthermore, for reliable wireless communication services, the quality of service (QoS) requirements of individual users are considered as a part of rewards and 60GHz mmWave radio is used for mobile access. This paper considers the 60GHz mmWave access for utilizing the benefits of i) ultra-wide-bandwidth for multi-Gbps high-speed communications and ii) high-directional communications for spatial reuse that is obviously good for densely deployed users. Lastly, the comprehensive and data-intensive performance evaluation of the proposed MADRL-based algorithm for multi-UAV positioning is conducted in this paper. The results of these evaluations demonstrate that the proposed algorithm outperforms other existing algorithms.

Notwithstanding its difficulty, forecasting of the development of intricate non-linear dynamical systems has been in a spotlight of various scientific fields (Strogatz, 2001; Jackson and Radunskaya, 2015). A plausible approach to forecasting the development is to isolate the non-linear estimation problem into (i) learning non-linear representations by applying highly non-linear functions such as deep neural networks (Goodfellow et al., 2016), and (ii) estimating its development with simple linear models. An example is a reservoir computing (RC; Jaeger, 2001, 2002). RC first randomly specifies a state in the reservoir layer in recurrent neural network (Rumelhart et al., 1986), and optimizes the weights only in the output layer; RC corresponds to non-linearly transform its input (in the reservoir layer) and trains a simple linear prediction model (in the output layer). It has been reported that such a simple combination of the non-linear representation learning and the linear estimation is effective to forecasting the intricate dynamical systems (Tanaka et al., 2019). Effectiveness of the simple combination is not limited to RC; applying a linear model to the non-linear representation in more general deep neural network is also regarded as a solid forecasting strategy (Lusch et al., 2018). Unfortunately, however, partial degrees of freedom corresponding to several state variables are not observed in some practical situations (Lucor et al., 2022; Cheng et al., 2023). There could be a variety of reasons for missing observations: it would be caused by the difficulty of measurement, it would be caused by the immature understanding of the system of interest, and so forth.

In recent years, the intelligence of various parts of the home has become one of the essential features of any modern home. One of these parts is the intelligence lighting system that personalizes the light for each person. This paper proposes an intelligent system based on machine learning that personalizes lighting in the instant future location of a recognized user, inferred by trajectory prediction. Our proposed system consists of the following modules: (I) human detection to detect and localize the person in each given video frame, (II) face recognition to identify the detected person, (III) human tracking to track the person in the sequence of video frames and (IV) trajectory prediction to forecast the future location of the user in the environment using Inverse Reinforcement Learning. The proposed method provides a unique profile for each person, including specifications, face images, and custom lighting settings. This profile is used in the lighting adjustment process. Unlike other methods that consider constant lighting for every person, our system can apply each 'person's desired lighting in terms of color and light intensity without direct user intervention. Therefore, the lighting is adjusted with higher speed and better efficiency. In addition, the predicted trajectory path makes the proposed system apply the desired lighting, creating more pleasant and comfortable conditions for the home residents. In the experimental results, the system applied the desired lighting in an average time of 1.4 seconds from the moment of entry, as well as a performance of 22.1mAp in human detection, 95.12% accuracy in face recognition, 93.3% MDP in human tracking, and 10.80 MinADE20, 18.55 MinFDE20, 15.8 MinADE5 and 30.50 MinFDE5 in trajectory prediction.

Probability and moment inequalities for sums of random variables are of paramount importance in the complexity analysis of numerous stochastic approximation algorithms or finite-time analysis of Monte Carlo estimators; see [20], [10], and references therein. The main focus in this area has been on concentration inequalities for independent random variable sums or martingale difference sequences; see e.g. in [4, 36]. However, the study of concentration inequalities for additive Markov chain functions is still relatively underdeveloped. For the technically simple case of uniformly ergodic Markov chains, there is extensive work on Hoeffding-and Bernstein-like inequalities as found in [23, 34, 20, 38]. Nevertheless, the application of these results may be difficult due to a lack of quantitative data or the substitution of asymptotic variance of the chain by surrogates; see Section 2.1 for relevant definitions. The present work aims to fill this gap by extending Rosenthal-and Bernstein-type inequalities to Markov chains which converge geometrically fast to a unique invariant distribution, with an explicit emphasis on the mixing time of the underlying Markov chain. An important tool for establishing deviation bounds for sums of random variables is based on moment inequalities.

For min-max optimization and variational inequalities problems (VIP) encountered in diverse machine learning tasks, Stochastic Extragradient (SEG) and Stochastic Gradient Descent Ascent (SGDA) have emerged as preeminent algorithms. Constant step-size variants of SEG/SGDA have gained popularity, with appealing benefits such as easy tuning and rapid forgiveness of initial conditions, but their convergence behaviors are more complicated even in rudimentary bilinear models. Our work endeavors to elucidate and quantify the probabilistic structures intrinsic to these algorithms. By recasting the constant step-size SEG/SGDA as time-homogeneous Markov Chains, we establish a first-of-its-kind Law of Large Numbers and a Central Limit Theorem, demonstrating that the average iterate is asymptotically normal with a unique invariant distribution for an extensive range of monotone and non-monotone VIPs. Specializing to convex-concave min-max optimization, we characterize the relationship between the step-size and the induced bias with respect to the Von-Neumann's value. Finally, we establish that Richardson-Romberg extrapolation can improve proximity of the average iterate to the global solution for VIPs. Our probabilistic analysis, underpinned by experiments corroborating our theoretical discoveries, harnesses techniques from optimization, Markov chains, and operator theory.