Hierarchical Deep Learning for Intention Estimation of Teleoperation Manipulation in Assembly Tasks

In human-robot collaboration, shared control presents an opportunity to teleoperate robotic manipulation to improve the efficiency of manufacturing and assembly processes. Robots are expected to assist in executing the user's intentions. To this end, robust and prompt intention estimation is needed, relying on behavioral observations. The framework presents an intention estimation technique at hierarchical levels i.e., low-level actions and high-level tasks, by incorporating multi-scale hierarchical information in neural networks. Technically, we employ hierarchical dependency loss to boost overall accuracy. Furthermore, we propose a multi-window method that assigns proper hierarchical prediction windows of input data. An analysis of the predictive power with various inputs demonstrates the predominance of the deep hierarchical model in the sense of prediction accuracy and early intention identification. We implement the algorithm on a virtual reality (VR) setup to teleoperate robotic hands in a simulation with various assembly tasks to show the effectiveness of online estimation.

Model-free Motion Planning of Autonomous Agents for Complex Tasks in Partially Observable Environments

Motion planning of autonomous agents in partially known environments with incomplete information is a challenging problem, particularly for complex tasks. This paper proposes a model-free reinforcement learning approach to address this problem. We formulate motion planning as a probabilistic-labeled partially observable Markov decision process (PL-POMDP) problem and use linear temporal logic (LTL) to express the complex task. The LTL formula is then converted to a limit-deterministic generalized B\"uchi automaton (LDGBA). The problem is redefined as finding an optimal policy on the product of PL-POMDP with LDGBA based on model-checking techniques to satisfy the complex task. We implement deep Q learning with long short-term memory (LSTM) to process the observation history and task recognition. Our contributions include the proposed method, the utilization of LTL and LDGBA, and the LSTM-enhanced deep Q learning. We demonstrate the applicability of the proposed method by conducting simulations in various environments, including grid worlds, a virtual office, and a multi-agent warehouse. The simulation results demonstrate that our proposed method effectively addresses environment, action, and observation uncertainties. This indicates its potential for real-world applications, including the control of unmanned aerial vehicles (UAVs).

Efficient LQR-CBF-RRT*: Safe and Optimal Motion Planning

Control Barrier Functions (CBF) are a powerful tool for designing safety-critical controllers and motion planners. The safety requirements are encoded as a continuously differentiable function that maps from state variables to a real value, in which the sign of its output determines whether safety is violated. In practice, the CBFs can be used to enforce safety by imposing itself as a constraint in a Quadratic Program (QP) solved point-wise in time. However, this approach costs computational resources and could lead to infeasibility in solving the QP. In this paper, we propose a novel motion planning framework that combines sampling-based methods with Linear Quadratic Regulator (LQR) and CBFs. Our approach does not require solving the QPs for control synthesis and avoids explicit collision checking during samplings. Instead, it uses LQR to generate optimal controls and CBF to reject unsafe trajectories. To improve sampling efficiency, we employ the Cross-Entropy Method (CEM) for importance sampling (IS) to sample configurations that will enhance the path with higher probability and store computed optimal gain matrices in a hash table to avoid re-computation during rewiring procedure. We demonstrate the effectiveness of our method on nonlinear control affine systems in simulation.

Learning Minimally-Violating Continuous Control for Infeasible Linear Temporal Logic Specifications

This paper explores continuous-time control synthesis for target-driven navigation to satisfy complex high-level tasks expressed as linear temporal logic (LTL). We propose a model-free framework using deep reinforcement learning (DRL) where the underlying dynamic system is unknown (an opaque box). Unlike prior work, this paper considers scenarios where the given LTL specification might be infeasible and therefore cannot be accomplished globally. Instead of modifying the given LTL formula, we provide a general DRL-based approach to satisfy it with minimal violation. %\mminline{Need to decide if we're comfortable calling these "guarantees" due to the stochastic policy. I'm not repeating this comment everywhere that says "guarantees" but there are multiple places.} To do this, we transform a previously multi-objective DRL problem, which requires simultaneous automata satisfaction and minimum violation cost, into a single objective. By guiding the DRL agent with a sampling-based path planning algorithm for the potentially infeasible LTL task, the proposed approach mitigates the myopic tendencies of DRL, which are often an issue when learning general LTL tasks that can have long or infinite horizons. This is achieved by decomposing an infeasible LTL formula into several reach-avoid sub-tasks with shorter horizons, which can be trained in a modular DRL architecture. Furthermore, we overcome the challenge of the exploration process for DRL in complex and cluttered environments by using path planners to design rewards that are dense in the configuration space. The benefits of the presented approach are demonstrated through testing on various complex nonlinear systems and compared with state-of-the-art baselines. The Video demonstration can be found on YouTube Channel:\url{https://youtu.be/jBhx6Nv224E}.

Learning Signal Temporal Logic through Neural Network for Interpretable Classification

Machine learning techniques using neural networks have achieved promising success for time-series data classification. However, the models that they produce are challenging to verify and interpret. In this paper, we propose an explainable neural-symbolic framework for the classification of time-series behaviors. In particular, we use an expressive formal language, namely Signal Temporal Logic (STL), to constrain the search of the computation graph for a neural network. We design a novel time function and sparse softmax function to improve the soundness and precision of the neural-STL framework. As a result, we can efficiently learn a compact STL formula for the classification of time-series data through off-the-shelf gradient-based tools. We demonstrate the computational efficiency, compactness, and interpretability of the proposed method through driving scenarios and naval surveillance case studies, compared with state-of-the-art baselines.

A Robotic Visual Grasping Design: Rethinking Convolution Neural Network with High-Resolutions

High-resolution representations are important for vision-based robotic grasping problems. Existing works generally encode the input images into low-resolution representations via sub-networks and then recover high-resolution representations. This will lose spatial information, and errors introduced by the decoder will be more serious when multiple types of objects are considered or objects are far away from the camera. To address these issues, we revisit the design paradigm of CNN for robotic perception tasks. We demonstrate that using parallel branches as opposed to serial stacked convolutional layers will be a more powerful design for robotic visual grasping tasks. In particular, guidelines of neural network design are provided for robotic perception tasks, e.g., high-resolution representation and lightweight design, which respond to the challenges in different manipulation scenarios. We then develop a novel grasping visual architecture referred to as HRG-Net, a parallel-branch structure that always maintains a high-resolution representation and repeatedly exchanges information across resolutions. Extensive experiments validate that these two designs can effectively enhance the accuracy of visual-based grasping and accelerate network training. We show a series of comparative experiments in real physical environments at Youtube: https://youtu.be/Jhlsp-xzHFY.

Overcoming Exploration: Deep Reinforcement Learning in Complex Environments from Temporal Logic Specifications

We present a Deep Reinforcement Learning (DRL) algorithm for a task-guided robot with unknown continuous-time dynamics deployed in a large-scale complex environment. Linear Temporal Logic (LTL) is applied to express a rich robotic specification. To overcome the environmental challenge, we propose a novel path planning-guided reward scheme that is dense over the state space, and crucially, robust to infeasibility of computed geometric paths due to the unknown robot dynamics. To facilitate LTL satisfaction, our approach decomposes the LTL mission into sub-tasks that are solved using distributed DRL, where the sub-tasks are trained in parallel, using Deep Policy Gradient algorithms. Our framework is shown to significantly improve performance (effectiveness, efficiency) and exploration of robots tasked with complex missions in large-scale complex environments.

Time-Incremental Learning from Data Using Temporal Logics

Real-time and human-interpretable decision-making in cyber-physical systems is a significant but challenging task, which usually requires predictions of possible future events from limited data. In this paper, we introduce a time-incremental learning framework: given a dataset of labeled signal traces with a common time horizon, we propose a method to predict the label of a signal that is received incrementally over time, referred to as prefix signal. Prefix signals are the signals that are being observed as they are generated, and their time length is shorter than the common horizon of signals. We present a novel decision-tree based approach to generate a finite number of Signal Temporal Logic (STL) specifications from the given dataset, and construct a predictor based on them. Each STL specification, as a binary classifier of time-series data, captures the temporal properties of the dataset over time. The predictor is constructed by assigning time-variant weights to the STL formulas. The weights are learned by using neural networks, with the goal of minimizing the misclassification rate for the prefix signals defined over the given dataset. The learned predictor is used to predict the label of a prefix signal, by computing the weighted sum of the robustness of the prefix signal with respect to each STL formula. The effectiveness and classification performance of our algorithm are evaluated on an urban-driving and a naval-surveillance case studies.

Intelligent Traffic Light via Policy-based Deep Reinforcement Learning

Intelligent traffic lights in smart cities can optimally reduce traffic congestion. In this study, we employ reinforcement learning to train the control agent of a traffic light on a simulator of urban mobility. As a difference from existing works, a policy-based deep reinforcement learning method, Proximal Policy Optimization (PPO), is utilized other than value-based methods such as Deep Q Network (DQN) and Double DQN (DDQN). At first, the obtained optimal policy from PPO is compared to those from DQN and DDQN. It is found that the policy from PPO performs better than the others. Next, instead of the fixed-interval traffic light phases, we adopt the light phases with variable time intervals, which result in a better policy to pass the traffic flow. Then, the effects of environment and action disturbances are studied to demonstrate the learning-based controller is robust. At last, we consider unbalanced traffic flows and find that an intelligent traffic light can perform moderately well for the unbalanced traffic scenarios, although it learns the optimal policy from the balanced traffic scenarios only.

Online Motion Planning with Soft Timed Temporal Logic in Dynamic and Unknown Environment

Motion planning of an autonomous system with high-level specifications has wide applications. However, research of formal languages involving timed temporal logic is still under investigation. Furthermore, many existing results rely on a key assumption that user-specified tasks are feasible in the given environment. Challenges arise when the operating environment is dynamic and unknown since the environment can be found prohibitive, leading to potentially conflicting tasks where pre-specified timed missions cannot be fully satisfied. Such issues become even more challenging when considering timed requirements. To address these challenges, this work proposes a control framework that considers hard constraints to enforce safety requirements and soft constraints to enable task relaxation. The metric interval temporal logic (MITL) specifications are employed to deal with time constraints. By constructing a relaxed timed product automaton, an online motion planning strategy is synthesized with a receding horizon controller to generate policies, achieving multiple objectives in decreasing order of priority 1) formally guarantee the satisfaction of hard safety constraints; 2) mostly fulfill soft timed tasks; and 3) collect time-varying rewards as much as possible. Another novelty of the relaxed structure is to consider violations of both time and tasks for infeasible cases. Simulation results are provided to validate the proposed approach.