IROS 2023 论文列表

Quadruped robots have the potential to guide blind and low vision (BLV) people due to their highly flexible locomotion and emotional value provided by their bionic forms. However, the development of quadruped guide robots rarely involves BLV users' participatory designs and evaluations. In this paper, we conducted two empirical experiments both in indoor controlled and outdoor field scenarios, exploring the benefits and drawbacks of quadruped guide robots. The results show that the nowadays commercial quadruped robots exposed significant disadvantages in usability and trust compared with wheeled robots. It is concluded that the moving gait and walking noise of quadruped robots would limit the guiding effectiveness to a certain extent, and the empathetic effect of its bionic form for BLV users could not be fully reflected. Based on the findings of wheeled robots and quadruped robots' advantages, we discuss the design implications for the future guide robot design for BLV users. This paper reports the first empirical experiment about quadruped guide robots with BLV users and preliminary explores their potential improvement space in substituting guide dogs, which can inspire the further specialized design of quadruped guide robots.

Effective BEV object detection on infrastructure can greatly improve traffic scene understanding and vehicle-to-infrastructure (V2I) cooperative perception. However, cameras installed on infrastructure have various postures, and previous BEV detection methods rely on accurate calibration, which is difficult for practical applications due to inevitable natural factors (e.g., wind and snow). In this paper, we propose a Calibration-free BEV Representation (CBR) network, which achieves 3D detection based on BEV representation without calibration parameters and additional depth supervision. Specifically, we utilize two multi-layer perceptrons for decoupling the features from perspective view to front view and bird-eye view under boxes-induced foreground supervision. Then, a cross-view feature fusion module matches features from orthogonal views according to similarity and conducts BEV feature enhancement with front-view features. Experimental results on DAIR-V2X demonstrate that CBR achieves acceptable performance without any camera parameters and is naturally not affected by calibration noises. We hope CBR can serve as a baseline for future research addressing practical challenges of infrastructure perception.

Phase variable control based on global tibia kinematics holds promise for predicting gait cycle progression to continuously control robotic transtibial prostheses. Calibration of the phase variable is critical to ensure its monotonic behavior, to approach a linear relationship with gait percentage, and to accurately predict the percentage of gait. This paper compares four calibration approaches using data from 22 able-bodied subjects walking at 14 speeds [1]. The typical pure centering (PC) approach employed for thigh-based phase variables is not viable, yielding monotonic phase progression in fewer than half of the cases. An optimization (OPT) approach found monotonic calibrations in 305/308 cases with high linearity (average R2 of 0.91). Critical point centering (CPC) approximates the OPT performance, with 274/308 monotonic calibrations and an average R2 of 0.85, whereas the related vertical weighted average (VWA) approach was only slightly better than PC. All four approaches are similarly accurate in predicting gait percentage, staying within 5% at least 92.7% of the time.

Mutual localization plays a crucial role in multi-robot cooperation. CREPES, a novel system that focuses on six degrees of freedom (DOF) relative pose estimation for multi-robot systems, is proposed in this paper. CREPES has a compact hardware design using active infrared (IR) LEDs, an IR fish-eye camera, an ultra-wideband (UWB) module and an inertial measurement unit (IMU). By leveraging IR light communication, the system solves data association between visual detection and UWB ranging. Ranging measurements from the UWB and directional information from the camera offer relative 3-DOF position estimation. Combining the mutual relative position with neighbors and the gravity constraints provided by IMUs, we can estimate the 6-DOF relative pose from a single frame of sensor measurements. In addition, we design an estimator based on the error-state Kalman filter (ESKF) to enhance system accuracy and robustness. When multiple neighbors are available, a Pose Graph Optimization (PGO) algorithm is applied to further improve system accuracy. We conduct enormous experiments to demonstrate CREPES’ accuracy between robot pairs and a team of robots, as well as performance under challenging conditions.

Handling fragile objects requires minimally invasive interaction skills in order to avoid any permanent deformation, alternation or damages. Such need is often required in tactile exploration tasks. In this paper, we propose an innovative whisker manipulator (COMPlacent), which is designed to accomplish tactile exploration with minimum intrusiveness. The design is inspired by biological whiskers observed in animals, where whiskers are used as means of tactile exploration in an analogous way as fingers are. Artificial whiskers are compliant but robust, which mitigates contact forces by bending or conforming to the object surface. The intrusiveness is further reduced by reactive control, which is implemented based on tactile sensors and actuators installed on each whisker. This allows the whisker to be retracted from the object surface, so that the energy transferred by contacts is minimized. The tactile sensor is designed to be ultrasensitive, which allows it to gather contact information with high fidelity. By modeling contact pressure as a time-series signal, a machine learning framework is leveraged to discriminate object properties including shape and texture. Evaluation experiments were conducted on real objects, which successfully demonstrates object classification at an accuracy of 97.3%, and texture discrimination accuracy of 92.1%.

Human motion prediction is important for mobile service robots and intelligent vehicles to operate safely and smoothly around people. The more accurate predictions are, particularly over extended periods of time, the better a system can, e.g., assess collision risks and plan ahead. In this paper, we propose to exploit maps of dynamics (MoDs, a class of general representations of place-dependent spatial motion patterns, learned from prior observations) for long-term human motion prediction (LHMP). We present a new MoD-informed human motion prediction approach, named CLiFF-LHMP, which is data efficient, explainable, and insensitive to errors from an upstream tracking system. Our approach uses CLiFF -map, a specific MoD trained with human motion data recorded in the same environment. We bias a constant velocity prediction with samples from the CLiFF-map to generate multi-modal trajectory predictions. In two public datasets we show that this algorithm outperforms the state of the art for predictions over very extended periods of time, achieving 45 % more accurate prediction performance at 50s compared to the baseline.

This paper presents a reactive planning system that allows a Cassie-series bipedal robot to avoid multiple non-overlapping obstacles via a single, continuously differentiable control barrier function (CBF). The overall system detects an individual obstacle via a height map derived from a LiDAR point cloud and computes an elliptical outer approximation, which is then turned into a CBF. The QP-CLF-CBF formalism developed by Ames et al. is applied to ensure that safe trajectories are generated. Safe planning in environments with multiple obstacles is demonstrated both in simulation and experimentally on the Cassie biped.

Human-robot interaction (HRI) is a rapidly growing field that encompasses social and industrial applications. Machine learning plays a vital role in industrial HRI by enhancing the adaptability and autonomy of robots in complex environments. However, data privacy is a crucial concern in the interaction between humans and robots, as companies need to protect sensitive data while machine learning algorithms require access to large datasets. Federated Learning (FL) offers a solution by enabling the distributed training of models without sharing raw data. Despite extensive research on Federated learning (FL) for tasks such as natural language processing (NLP) and image classification, the question of how to use FL for HRI remains an open research problem. The traditional FL approach involves transmitting large neural network parameter matrices between the server and clients, which can lead to high communication costs and often becomes a bottleneck in FL. This paper proposes a communication-efficient FL framework for human-robot interaction (CEFHRI) to address the challenges of data heterogeneity and communication costs. The framework leverages pre-trained models and introduces a trainable spatiotemporal adapter for video understanding tasks in HRI. Experimental results on three human-robot interaction benchmark datasets: HRI30, InHARD, and COIN demonstrate the superiority of CEFHRI over full fine-tuning in terms of communication costs. The proposed methodology provides a secure and efficient approach to HRI federated learning, particularly in industrial environments with data privacy concerns and limited communication bandwidth. Our code is available at https://github.com/umarkhalidAI/CEFHRI-Efficient-Federated-Learning.

The Dijkstra algorithm is a classic path planning method, which in a discrete graph space, can start from a specified source node and find the shortest path between the source node and all other nodes in the graph. However, to the best of our knowledge, there is no effective method that achieves a function similar to that of the Dijkstra's algorithm in a continuous space. In this study, an optimal path planning algorithm called convex dissection topology (CDT)-Dijkstra is developed, which can quickly compute the global optimal path from one point to all other points in a 2D continuous space. CDT-Dijkstra is mainly divided into two stages: SetInit and GetGoal. In SetInit, the algorithm can quickly obtain the optimal CDT encoding set of all the cut lines based on the initial point Xinit. In GetGoal, the algorithm can return the global optimal path of any goal point at an extremely high speed. In this study, we propose and prove the planning principle of considering only the points on the cutlines, thus reducing the state space of the distance optimal path planning task from 2D to 1D. In addition, we propose a fast method to find the optimal path in a homogeneous class and theoretically prove the correctness of the method. Finally, by testing in a series of environments, the experimental results demonstrate that CDT-Dijkstra not only plans the optimal path from all points at once, but also has a significant advantage over advanced algorithms considering certain complex tasks.

Current motion planning approaches rely on binary collision checking to evaluate the validity of a state and thereby dictate where the robot is allowed to move. This approach leaves little room for robots to engage in contact with an object, as is often necessary when operating in densely cluttered spaces. In this work, we propose an alternative method that considers contact states as high-cost states that the robot should avoid but can traverse if necessary to complete a task. More specifically, we introduce Contact Admissible Transition-based Rapidly exploring Random Trees (CAT-RRT)11Supplementary video and open source code [1]., a planner that uses a novel per-link cost heuristic to find a path by traversing high-cost obstacle regions. Through extensive testing, we find that state-of-the-art optimization planners tend to over-explore low-cost states, which leads to slow and inefficient convergence to contact regions. Conversely, CAT-RRT searches both low and high-cost regions simultaneously with an adaptive thresholding mechanism carried out at each robot link. This leads to paths with a balance between efficiency, path length, and contact cost.

Robots operating in real-world environments must reason about possible outcomes of stochastic actions and make decisions based on partial observations of the true world state. A major challenge for making accurate and robust action predictions is the problem of confounding, which if left untreated can lead to prediction errors. The partially observable Markov decision process (POMDP) is a widely-used framework to model these stochastic and partially-observable decision-making problems. However, due to a lack of explicit causal semantics, POMDP planning methods are prone to confounding bias and thus in the presence of unobserved confounders may produce underperforming policies. This paper presents a novel causally-informed extension of “anytime regularized determinized sparse partially observable tree” (AR-DESPOT), a modern anytime online POMDP planner, using causal modelling and inference to eliminate errors caused by unmeasured confounder variables. We further propose a method to learn offline the partial parameterisation of the causal model for planning, from ground truth model data. We evaluate our methods on a toy problem with an unobserved confounder and show that the learned causal model is highly accurate, while our planning method is more robust to confounding and produces overall higher performing policies than AR-DESPOT.

This study presents the conflict-aware multi-agent estimated time of arrival (CAMETA) framework, a novel approach for predicting the arrival times of multiple agents in unstructured environments without predefined road infrastructure. The CAMETA framework consists of three components: a path planning layer generating potential path suggestions, a multi-agent ETA prediction layer predicting the arrival times for all agents based on the paths, and lastly, a path selection layer that calculates the accumulated cost and selects the best path. The novelty of the CAMETA framework lies in the heterogeneous map representation and the heterogeneous graph neural network architecture. As a result of the proposed novel structure, CAMETA improves the generalization capability compared to the state-of-the-art methods that rely on structured road infrastructure and historical data. The simulation results demonstrate the efficiency and efficacy of the multi-agent ETA prediction layer, with a mean average percentage error improvement of 29.5% and 44% when compared to a traditional path planning method (A *) which does not consider conflicts. The performance of the CAMETA framework shows significant improvements in terms of robustness to noise and conflicts as well as determining proficient routes compared to state-of-the-art multi-agent path planners.

With the increasing computing power, using data-driven approaches to co-design a robot's morphology and controller has become a promising way. However, most existing data-driven methods require training the controller for each morphology to calculate fitness, which is time-consuming. In contrast, the dual-network framework utilizes data collected by individual networks under a specific morphology to train a population network that provides a surrogate function for morphology optimization. This approach replaces the traditional evaluation of a diverse set of candidates, thereby speeding up the training. Despite considerable results, the online training of both networks impedes their performance. To address this issue, we propose a concurrent network framework that combines online and offline reinforcement learning (RL) methods. By leveraging the behavior cloning term in a flexible manner, we achieve an effective combination of both networks. We conducted multiple sets of comparative experiments in the simulator and found that the proposed method effectively addresses issues present in the dual-network framework, leading to overall algorithmic performance improvement. Furthermore, we validated the algorithm on a real robot, demonstrating its feasibility in a practical application.

We propose a mechanism for low Reynolds num-ber walking (e.g., legged microscale robots). Whereas locomotion for legged robots has traditionally been classified as dynamic (where inertia plays a role) or static (where the system is always statically stable), we introduce a new locomotion modality we call buoyancy enabled non-inertial dynamic walking in which inertia plays no role, yet the robot is not statically stable. Instead, falling and viscous drag play critical roles. This model assumes squeeze flow forces from fluid interactions combined with a well timed gait as the mechanism by which forward motion can be achieved from a reciprocating legged robot. Using two physical demonstrations of robots with Reynold's number ranging from 0.0001 to 0.02 (a microscale robot in water and a centimeter scale robot in glycerol) we find the model qualitatively describes the motion. This model can help understand microscale locomotion and design new microscale walking robots including controlling forward and backwards motion and potentially steering these robots.

Autonomous exploration is a crucial aspect of robotics that has numerous applications. Most of the existing methods greedily choose goals that maximize immediate reward. This strategy is computationally efficient but insufficient for overall exploration efficiency. In recent years, some state-of-the-art methods are proposed, which generate a global coverage path and significantly improve overall exploration efficiency. However, global optimization produces high computational overhead, leading to low-frequency planner updates and inconsistent planning motion. In this work, we propose a novel method to support fast UAV exploration in large-scale and cluttered 3-D environments. We introduce a computationally low-cost viewpoints generation method using occlusion-free spheres. Additionally, we combine greedy strategy with global optimization, which considers both computational and exploration efficiency. We benchmark our method against state-of-the-art methods to showcase its superiority in terms of exploration efficiency and computational time. We conduct various real-world experiments to demonstrate the excellent performance of our method in large-scale and cluttered environments.

The integration of semantic information can effectively enhance the performance of 3D object detection based on lidar point cloud. Most of previous researches utilize camera-lidar fusion to improve detection accuracy for distant or small objects. However, this approach is typically unsuitable for real-time applications due to the large amount of input data. Recently, a multi-task framework using only Iidar has emerged as an alternative that employs the same feature extraction backbone with different heads to simultaneously output detection and semantic segmentation results for lidar point clouds. Nonetheless, some previous works have failed to achieve an optimal balance between accuracy and speed. To address this issue, we propose a multi-task framework which leverages the Cartesian pillar and a multi-scale semantic segmentation head to overcome the shortcomings of existing works and improve the detection accuracy. We evaluate the proposed method using typical pillar-based and voxel-based detection models on the nuScenes dataset. The experimental results demonstrate that the proposed design achieves better performance especially on small objects, compared to single-task models. Moreover, the proposed network increases mAP and NDS by 3.1 % and 2.5 % respectively on the nuScenes test set, compared to the representative multi-task network.

Interactive reinforcement learning has shown promise in learning complex robotic tasks. However, the process can be human-intensive due to the requirement of a large amount of interactive feedback. This paper presents a new method that uses scores provided by humans instead of pairwise preferences to improve the feedback efficiency of interactive reinforcement learning. Our key insight is that scores can yield significantly more data than pairwise preferences. Specifically, we require a teacher to interactively score the full trajectories of an agent to train a behavioral policy in a sparse reward environment. To avoid unstable scores given by humans negatively impacting the training process, we propose an adaptive learning scheme. This enables the learning paradigm to be insensitive to imperfect or unreliable scores. We extensively evaluate our method for robotic locomotion and manipulation tasks. The results show that the proposed method can efficiently learn near-optimal policies by adaptive learning from scores while requiring less feedback compared to pairwise preference learning methods. The source codes are publicly available at https://github.com/SSKKai/Interactive-Scoring-IRL.

Robust, fast, and accurate human state - 6D pose and posture - estimation remains a challenging problem. For real-world applications, the ability to estimate the human state in realtime is highly desirable. In this paper, we present BodySLAM++, a fast, efficient, and accurate human and camera state estimation framework relying on visual-inertial data. BodySLAM++ extends an existing visual-inertial state estimation framework, OKVIS2, to solve the dual task of estimating camera and human states simultaneously. Our system improves the accuracy of both human and camera state estimation with respect to baseline methods by 26 % and 12%, respectively, and achieves realtime performance at 15+ frames per second on an Intel i7-model CPU. Experiments were conducted on a custom dataset containing both ground truth human and camera poses collected with an indoor motion tracking system.