A Dynamic Safety Shield for Safe and Efficient Reinforcement Learning of Navigation Tasks

Reinforcement learning (RL) has been successfully applied to a variety of robotics applications, where it outperforms classical methods. However, the safety aspect of RL and the transfer to the real world remain an open challenge. A prominent field for tackling this challenge and ensuring the safety of the agents during training and execution is safe reinforcement learning. Safe RL can be achieved through constrained RL and safe exploration approaches. The former learns the safety constraints over the course of training to achieve a safe behavior by the end of training, at the cost of high number of collisions at earlier stages of the training. The latter offers robust safety by enforcing the safety constraints as hard constraints, which prevents collisions but hinders the exploration of the RL agent, resulting in lower rewards and poor performance. To overcome those drawbacks, we propose a novel safety shield, that combines the robustness of the optimization-based controllers with the long prediction capabilities of the RL agents, allowing the RL agent to adaptively tune the parameters of the controller. Our approach is able to improve the exploration of the RL agents for navigation tasks, while minimizing the number of collisions. Experiments in simulation show that our approach outperforms state-of-the-art baselines in the reached goals-to-collisions ratio in different challenging environments. The goals-to-collisions ratio metrics emphasizes the importance of minimizing the number of collisions, while learning to accomplish the task. Our approach achieves a higher number of reached goals compared to the classic safety shields and fewer collisions compared to constrained RL approaches. Finally, we demonstrate the performance of the proposed method in a real-world experiment.

Safe Leaf Manipulation for Accurate Shape and Pose Estimation of Occluded Fruits

Fruit monitoring plays an important role in crop management, and rising global fruit consumption combined with labor shortages necessitates automated monitoring with robots. However, occlusions from plant foliage often hinder accurate shape and pose estimation. Therefore, we propose an active fruit shape and pose estimation method that physically manipulates occluding leaves to reveal hidden fruits. This paper introduces a framework that plans robot actions to maximize visibility and minimize leaf damage. We developed a novel scene-consistent shape completion technique to improve fruit estimation under heavy occlusion and utilize a perception-driven deformation graph model to predict leaf deformation during planning. Experiments on artificial and real sweet pepper plants demonstrate that our method enables robots to safely move leaves aside, exposing fruits for accurate shape and pose estimation, outperforming baseline methods. Project page: https://shaoxiongyao.github.io/lmap-ssc/.

Context-Based Meta Reinforcement Learning for Robust and Adaptable Peg-in-Hole Assembly Tasks

Peg-in-hole assembly in unknown environments is a challenging task due to onboard sensor errors, which result in uncertainty and variations in task parameters such as the hole position and orientation. Meta Reinforcement Learning (Meta RL) has been proposed to mitigate this problem as it learns how to quickly adapt to new tasks with different parameters. However, previous approaches either depend on a sample-inefficient procedure or human demonstrations to perform the task in the real world. Our work modifies the data used by the Meta RL agent and uses simple features that can be easily measured in the real world even with an uncalibrated camera. We further adapt the Meta RL agent to use data from a force/torque sensor, instead of the camera, to perform the assembly, using a small amount of training data. Finally, we propose a fine-tuning method that consistently and safely adapts to out-of-distribution tasks with parameters that differ by a factor of 10 from the training tasks. Our results demonstrate that the proposed data modification significantly enhances the training and adaptation efficiency and enables the agent to achieve 100% success in tasks with different hole positions and orientations. Experiments on a real robot confirm that both camera- and force/torque sensor-equipped agents achieve 100% success in tasks with unknown hole positions, matching their simulation performance and validating the approach's robustness and applicability. Compared to the previous work with sample-inefficient adaptation, our proposed methods are 10 times more sample-efficient in the real-world tasks.

Physically-Consistent Parameter Identification of Robots in Contact

Accurate inertial parameter identification is crucial for the simulation and control of robots encountering intermittent contact with the environment. Classically, robots' inertial parameters are obtained from CAD models that are not precise (and sometimes not available, e.g., Spot from Boston Dynamics), hence requiring identification. To do that, existing methods require access to contact force measurement, a modality not present in modern quadruped and humanoid robots. This paper presents an alternative technique that utilizes joint current/torque measurements -- a standard sensing modality in modern robots -- to identify inertial parameters without requiring direct contact force measurements. By projecting the whole-body dynamics into the null space of contact constraints, we eliminate the dependency on contact forces and reformulate the identification problem as a linear matrix inequality that can handle physical and geometrical constraints. We compare our proposed method against a common black-box identification mrethod using a deep neural network and show that incorporating physical consistency significantly improves the sample efficiency and generalizability of the model. Finally, we validate our method on the Spot quadruped robot across various locomotion tasks, showcasing its accuracy and generalizability in real-world scenarios over different gaits.

Constrained Object Placement Using Reinforcement Learning

Close and precise placement of irregularly shaped objects requires a skilled robotic system. Particularly challenging is the manipulation of objects that have sensitive top surfaces and a fixed set of neighbors. To avoid damaging the surface, they have to be grasped from the side, and during placement, their neighbor relations have to be maintained. In this work, we train a reinforcement learning agent that generates smooth end-effector motions to place objects as close as possible next to each other. During the placement, our agent considers neighbor constraints defined in a given layout of the objects while trying to avoid collisions. Our approach learns to place compact object assemblies without the need for predefined spacing between objects as required by traditional methods. We thoroughly evaluated our approach using a two-finger gripper mounted to a robotic arm with six degrees of freedom. The results show that our agent outperforms two baseline approaches in terms of object assembly compactness, thereby reducing the needed space to place the objects according to the given neighbor constraints. On average, our approach reduces the distances between all placed objects by at least 60%, with fewer collisions at the same compactness compared to both baselines.

Learning Adaptive Multi-Objective Robot Navigation with Demonstrations

Preference-aligned robot navigation in human environments is typically achieved through learning-based approaches, utilizing demonstrations and user feedback for personalization. However, personal preferences are subject to change and might even be context-dependent. Yet traditional reinforcement learning (RL) approaches with a static reward function often fall short in adapting to these varying user preferences. This paper introduces a framework that combines multi-objective reinforcement learning (MORL) with demonstration-based learning. Our approach allows for dynamic adaptation to changing user preferences without retraining. Through rigorous evaluations, including sim-to-real and robot-to-robot transfers, we demonstrate our framework's capability to reflect user preferences accurately while achieving high navigational performance in terms of collision avoidance and goal pursuance.

Sound Matters: Auditory Detectability of Mobile Robots

Mobile robots are increasingly being used in noisy environments for social purposes, e.g. to provide support in healthcare or public spaces. Since these robots also operate beyond human sight, the question arises as to how different robot types, ambient noise or cognitive engagement impacts the detection of the robots by their sound. To address this research gap, we conducted a user study measuring auditory detection distances for a wheeled (Turtlebot 2i) and quadruped robot (Unitree Go 1), which emit different consequential sounds when moving. Additionally, we also manipulated background noise levels and participants' engagement in a secondary task during the study. Our results showed that the quadruped robot sound was detected significantly better (i.e., at a larger distance) than the wheeled one, which demonstrates that the movement mechanism has a meaningful impact on the auditory detectability. The detectability for both robots diminished significantly as background noise increased. But even in high background noise, participants detected the quadruped robot at a significantly larger distance. The engagement in a secondary task had hardly any impact. In essence, these findings highlight the critical role of distinguishing auditory characteristics of different robots to improve the smooth human-centered navigation of mobile robots in noisy environments.

A Neuromorphic Approach to Obstacle Avoidance in Robot Manipulation

Neuromorphic computing mimics computational principles of the brain in $\textit{silico}$ and motivates research into event-based vision and spiking neural networks (SNNs). Event cameras (ECs) exclusively capture local intensity changes and offer superior power consumption, response latencies, and dynamic ranges. SNNs replicate biological neuronal dynamics and have demonstrated potential as alternatives to conventional artificial neural networks (ANNs), such as in reducing energy expenditure and inference time in visual classification. Nevertheless, these novel paradigms remain scarcely explored outside the domain of aerial robots. To investigate the utility of brain-inspired sensing and data processing, we developed a neuromorphic approach to obstacle avoidance on a camera-equipped manipulator. Our approach adapts high-level trajectory plans with reactive maneuvers by processing emulated event data in a convolutional SNN, decoding neural activations into avoidance motions, and adjusting plans using a dynamic motion primitive. We conducted experiments with a Kinova Gen3 arm performing simple reaching tasks that involve obstacles in sets of distinct task scenarios and in comparison to a non-adaptive baseline. Our neuromorphic approach facilitated reliable avoidance of imminent collisions in simulated and real-world experiments, where the baseline consistently failed. Trajectory adaptations had low impacts on safety and predictability criteria. Among the notable SNN properties were the correlation of computations with the magnitude of perceived motions and a robustness to different event emulation methods. Tests with a DAVIS346 EC showed similar performance, validating our experimental event emulation. Our results motivate incorporating SNN learning, utilizing neuromorphic processors, and further exploring the potential of neuromorphic methods.

EnQuery: Ensemble Policies for Diverse Query-Generation in Preference Alignment of Robot Navigation

To align mobile robot navigation policies with user preferences through reinforcement learning from human feedback (RLHF), reliable and behavior-diverse user queries are required. However, deterministic policies fail to generate a variety of navigation trajectory suggestions for a given navigation task configuration. We introduce EnQuery, a query generation approach using an ensemble of policies that achieve behavioral diversity through a regularization term. For a given navigation task, EnQuery produces multiple navigation trajectory suggestions, thereby optimizing the efficiency of preference data collection with fewer queries. Our methodology demonstrates superior performance in aligning navigation policies with user preferences in low-query regimes, offering enhanced policy convergence from sparse preference queries. The evaluation is complemented with a novel explainability representation, capturing full scene navigation behavior of the mobile robot in a single plot.

Learning Goal-Directed Object Pushing in Cluttered Scenes with Location-Based Attention

Non-prehensile planar pushing is a challenging task due to its underactuated nature with hybrid-dynamics, where a robot needs to reason about an object's long-term behaviour and contact-switching, while being robust to contact uncertainty. The presence of clutter in the environment further complicates this task, introducing the need to include more sophisticated spatial analysis to avoid collisions. Building upon prior work on reinforcement learning (RL) with multimodal categorical exploration for planar pushing, in this paper we incorporate location-based attention to enable robust navigation through clutter. Unlike previous RL literature addressing this obstacle avoidance pushing task, our framework requires no predefined global paths and considers the target orientation of the manipulated object. Our results demonstrate that the learned policies successfully navigate through a wide range of complex obstacle configurations, including dynamic obstacles, with smooth motions, achieving the desired target object pose. We also validate the transferability of the learned policies to robotic hardware using the KUKA iiwa robot arm.