RL-augmented MPC Framework for Agile and Robust Bipedal Footstep Locomotion Planning and Control Seung Hyeon Bang 1, Carlos Arribalzaga Jov e 1, 2, and Luis Sentis 1 Abstract -- This paper proposes an online bipedal footstep planning strategy that combines model predictive control (MPC) and reinforcement learning (RL) to achieve agile and robust bipedal maneuvers. While MPC-based foot placement controllers have demonstrated their effectiveness in achieving dynamic locomotion, their performance is often limited by the use of simplified models and assumptions. T o address this challenge, we develop a novel foot placement controller that leverages a learned policy to bridge the gap between the use of a simplified model and the more complex full-order robot system. Specifically, our approach employs a unique combination of an ALIP-based MPC foot placement controller for sub-optimal footstep planning and the learned policy for refining footstep adjustments, enabling the resulting footstep policy to capture the robot's whole-body dynamics effectively. We validate the effectiveness of our framework through a series of experiments using the full-body humanoid robot DRACO 3. The results demonstrate significant improvements in dynamic locomotion performance, including better tracking of a wide range of walking speeds, enabling reliable turning and traversing challenging terrains while preserving the robustness and stability of the walking gaits compared to the baseline ALIP-based MPC approach. I. INTRODUCTION Agile and robust bipedal locomotion is essential for humanoid robots to achieve human-level performance. One of the main challenges in achieving this is designing a footstep policy that enables bipeds to constantly adjust their planned footstep positions to maintain balance as well as to achieve more agile and fast maneuvers, even while traversing adverse environments, such as external disturbances or challenging terrains. In this paper, we present an RL-augmented MPC framework designed to generate a footstep policy for agile and robust bipedal locomotion.

With the escalating demand for meal delivery services, this study delves into the Meal Delivery Routing Problem (MDRP) within the context of last-mile logis-tics. Focusing on the critical aspects of courier allocation and order fulfillment, we introduce a novel approach utilizing a GRASP metaheuristic. The algorithm optimizes the assignment of couriers to orders, considering dynamic factors such as courier availability, order demands, and geographical locations. Real-world in-stances from a Colombian delivery app form the basis of our computational anal-ysis. Calibration of GRASP parameters reveals a delicate trade-off between solu-tion quality and computational time. Comparative results with a simulation-optimization based study underscore GRASP's competitive performance, demon-strating strengths in fulfilling orders and routing efficiency across diverse in-stances. This research enhances operational efficiency in the burgeoning food de-livery industry, shedding light on practical algorithms for last-mile logistics opti-mization.

With recent advancements in AI and computation tools, intelligent paradigms emerged to empower different fields such as healthcare robots with new capabilities. Advanced AI robotic algorithms (e.g., reinforcement learning) can be trained and developed to autonomously make individual decisions to achieve a desired and usually fixed goal. However, such independent decisions and goal achievements might not be ideal for a healthcare robot that usually interacts with a dynamic end-user or a patient. In such a complex human-robot interaction (teaming) framework, the dynamic user continuously wants to be involved in decision-making as well as introducing new goals while interacting with their present environment in real - time. T o address this challenge, an adaptive shared autonomy AI paradigm is required to be developed for the two interactive agents (Human & AI agents) w ith a foundation based on human-centered factors to avoid any possible ethical issues and guarantee no harm to humanity.

Over the years, complex control approaches have been developed to control the motion of a bicycle. Reinforcement Learning (RL), a branch of machine learning, promises easy deployment of so-called agents. Deployed agents are increasingly considered as an alternative to controllers for mechanical systems. The present work introduces an RL approach to do path following with a virtual bicycle model while simultaneously stabilising it laterally. The bicycle, modelled as the Whipple benchmark model and using multibody system dynamics, has no stabilisation aids. The agent succeeds in both path following and stabilisation of the bicycle model exclusively by outputting steering angles, which are converted into steering torques via a PD controller. Curriculum learning is applied as a state-of-the-art training strategy. Different settings for the implemented RL framework are investigated and compared to each other. The performance of the deployed agents is evaluated using different types of paths and measurements. The ability of the deployed agents to do path following and stabilisation of the bicycle model travelling between 2m/s and 7m/s along complex paths including full circles, slalom manoeuvres, and lane changes is demonstrated. Explanatory methods for machine learning are used to analyse the functionality of a deployed agent and link the introduced RL approach with research in the field of bicycle dynamics.

Despite the availability of international prize-money competitions, scaled vehicles, and simulation environments, research on autonomous racing and the control of sports cars operating close to the limit of handling has been limited by the high costs of vehicle acquisition and management, as well as the limited physics accuracy of open-source simulators. In this paper, we propose a racing simulation platform based on the simulator Assetto Corsa to test, validate, and benchmark autonomous driving algorithms, including reinforcement learning (RL) and classical Model Predictive Control (MPC), in realistic and challenging scenarios. Our contributions include the development of this simulation platform, several state-of-the-art algorithms tailored to the racing environment, and a comprehensive dataset collected from human drivers. Additionally, we evaluate algorithms in the offline RL setting.

We study reinforcement learning (RL) for a class of continuous-time linear-quadratic (LQ) control problems for diffusions where volatility of the state processes depends on both state and control variables. We apply a model-free approach that relies neither on knowledge of model parameters nor on their estimations, and devise an actor-critic algorithm to learn the optimal policy parameter directly. Our main contributions include the introduction of a novel exploration schedule and a regret analysis of the proposed algorithm. We provide the convergence rate of the policy parameter to the optimal one, and prove that the algorithm achieves a regret bound of $O(N^{\frac{3}{4}})$ up to a logarithmic factor. We conduct a simulation study to validate the theoretical results and demonstrate the effectiveness and reliability of the proposed algorithm. We also perform numerical comparisons between our method and those of the recent model-based stochastic LQ RL studies adapted to the state- and control-dependent volatility setting, demonstrating a better performance of the former in terms of regret bounds.

Visual reinforcement learning (RL) has made significant progress in recent years, but the choice of visual feature extractor remains a crucial design decision. This paper compares the performance of RL algorithms that train a convolutional neural network (CNN) from scratch with those that utilize pre-trained visual representations (PVRs). We evaluate the Dormant Ratio Minimization (DRM) algorithm, a state-of-the-art visual RL method, against three PVRs: ResNet18, DINOv2, and Visual Cortex (VC). We use the Metaworld Push-v2 and Drawer-Open-v2 tasks for our comparison. Our results show that the choice of training from scratch compared to using PVRs for maximising performance is task-dependent, but PVRs offer advantages in terms of reduced replay buffer size and faster training times. We also identify a strong correlation between the dormant ratio and model performance, highlighting the importance of exploration in visual RL. Our study provides insights into the trade-offs between training from scratch and using PVRs, informing the design of future visual RL algorithms.

This paper introduces MoveLight, a novel traffic signal control system that enhances urban traffic management through movement-centric deep reinforcement learning. By leveraging detailed real-time data and advanced machine learning techniques, MoveLight overcomes the limitations of traditional traffic signal control methods. It employs a lane-level control approach using the FRAP algorithm to achieve dynamic and adaptive traffic signal control, optimizing traffic flow, reducing congestion, and improving overall efficiency. Our research demonstrates the scalability and effectiveness of MoveLight across single intersections, arterial roads, and network levels. Experimental results using real-world datasets from Cologne and Hangzhou show significant improvements in metrics such as queue length, delay, and throughput compared to existing methods. This study highlights the transformative potential of deep reinforcement learning in intelligent traffic signal control, setting a new standard for sustainable and efficient urban transportation systems.

Reinforcement Learning (RL) has enabled social robots to generate trajectories without human-designed rules or interventions, which makes it more effective than hard-coded systems for generalizing to complex real-world scenarios. However, social navigation is a safety-critical task that requires robots to avoid collisions with pedestrians while previous RL-based solutions fall short in safety performance in complex environments. To enhance the safety of RL policies, to the best of our knowledge, we propose the first algorithm, SoNIC, that integrates adaptive conformal inference (ACI) with constrained reinforcement learning (CRL) to learn safe policies for social navigation. More specifically, our method augments RL observations with ACI-generated nonconformity scores and provides explicit guidance for agents to leverage the uncertainty metrics to avoid safety-critical areas by incorporating safety constraints with spatial relaxation. Our method outperforms state-of-the-art baselines in terms of both safety and adherence to social norms by a large margin and demonstrates much stronger robustness to out-of-distribution scenarios. Our code and video demos are available on our project website: https://sonic-social-nav.github.io/.

This paper investigates multi-objective reinforcement learning (MORL), which focuses on learning Pareto optimal policies in the presence of multiple reward functions. Despite MORL's significant empirical success, there is still a lack of satisfactory understanding of various MORL optimization targets and efficient learning algorithms. Our work offers a systematic analysis of several optimization targets to assess their abilities to find all Pareto optimal policies and controllability over learned policies by the preferences for different objectives. We then identify Tchebycheff scalarization as a favorable scalarization method for MORL. Considering the non-smoothness of Tchebycheff scalarization, we reformulate its minimization problem into a new min-max-max optimization problem. Then, for the stochastic policy class, we propose efficient algorithms using this reformulation to learn Pareto optimal policies. We first propose an online UCB-based algorithm to achieve an $\varepsilon$ learning error with an $\tilde{\mathcal{O}}(\varepsilon^{-2})$ sample complexity for a single given preference. To further reduce the cost of environment exploration under different preferences, we propose a preference-free framework that first explores the environment without pre-defined preferences and then generates solutions for any number of preferences. We prove that it only requires an $\tilde{\mathcal{O}}(\varepsilon^{-2})$ exploration complexity in the exploration phase and demands no additional exploration afterward. Lastly, we analyze the smooth Tchebycheff scalarization, an extension of Tchebycheff scalarization, which is proved to be more advantageous in distinguishing the Pareto optimal policies from other weakly Pareto optimal policies based on entry values of preference vectors. Furthermore, we extend our algorithms and theoretical analysis to accommodate this optimization target.