Autonomous Mobile Systems Research Articles

Shared low-speed autonomous vehicles for short-distance trips: agent-based modeling with mode choice analysis

ABSTRACT Previous research has envisioned behavior changes with the availability of shared autonomous vehicle (SAV) services. It is expected that SAVs will compete with conventional human-driven vehicles and lead to a shift in mode share. If designed as single-occupancy, low-speed, and lightweight, SAVs can also be considered a micro-mobility mode targeting short-distance trips in geofencing areas, which has been lacking in discussion in the state-of-the-art literature. To address this gap, this study envisions changes in mode choice with the implementation of a single-occupancy Shared Low-speed Light-weight Autonomous Mobility (SLLAM) system deployed, serving trips at speeds below 25 mph. Agent-based simulations are conducted to evaluate the mode share of the SLLAM system. The findings demonstrate that the SLLAM system has the potential to serve 36.98% of the trip demand in the study area. Increased operating speeds enhance the SLLAM mode share, making it a more attractive option for existing private vehicle trips.

Special Issue on Control and Applications of Multi-Agent Systems

“Multi-agent systems (MAS)” have been extensively studied across various fields, including robotics, economics, biology, and computer science. A distinctive feature of these systems is the ability of multiple agents, each with different characteristics, to perform system-wide tasks through local bottom-up interactions. Furthermore, design and control methods for system networks based on graph theory are being developed. Recent applications of these methods include autonomous driving technology, smart grids, and understanding social systems. This special issue aims to deepen the understanding of MAS, focusing on their control and applications. It features 16 papers, including one review paper. The accepted papers cover a wide range of topics, including reinforcement learning, autonomous mobility systems, and machine learning, presenting the latest research findings on MAS. These studies provide valuable insights into various aspects and potential applications of MAS. We hope that this issue will be beneficial to our readers and contribute to the advancement of future research.

Participatory Planning in Designing Autonomous Mobility Systems: A Design Thinking Approach

The adoption of participatory planning approaches in the deployment of new mobility solutions is rare. In this study, we conducted a community-based planning exercise using design thinking approaches to engage residents in co-creating an Autonomous Mobility System (AMS) for their neighborhood in Singapore. Through this process, a baseline understanding of participants’ attitude toward autonomous vehicles (AVs) was established, along with understanding the residents’ travel patterns and challenges. A co-creation activity guided participants to design their ideal AMS for the local community living in the Punggol residential district in Singapore. The exercise resulted in 20 unique designs of AMS, producing novel ideas such as suitable AV-holding areas and service zones. The results of the design activities provided insights into community-level priority and values. For AMS, providing alternative transport options that complemented existing train and bus services was the priority. As participants co-designed an AMS that fitted their needs as well as the needs of other participants, the designs also suggested that participants valued transport equity and accessibility, especially for the marginalized population, highlighting the importance of a human-centric design approach to transportation planning.

Performance estimation and operating policies in a truck-based autonomous mobile robot delivery system

Last-mile delivery is a costly and time-consuming process in e-commerce operations. Piloted by several companies, autonomous mobile robots cooperating with trucks have been implemented for parcel delivery. In this system, a truck departs from the warehouse, carrying robots and parcels, and traverses several drop-off points to release robots for parcel delivery or to deliver parcels by itself. We focus on performance estimation and system configuration, considering the no-zoning policy, where all trucks serve the whole delivery area and the zoning policy, where each truck takes charge of one zone. A queuing network is constructed to estimate the parcel throughput time and a cost minimisation model is developed for the system with a required throughput time. The optimal number of drop-off points and batch size can be numerically found to minimise the order throughput time. The zoning policy is cheaper than no-zoning policy for a small number of drop-off points and a short required throughput time, and it is cheaper in a system with a large number of drop-off points. Moreover, we find that while the truck-based autonomous mobile robot delivery system typically has a larger order throughput time than a delivery system using only robots, it is much less costly.

A bio-inspired model for robust navigation assistive devices

This paper proposes a new implementation and evaluation in a real-world environment of a bio-inspired predictive navigation model for mobility control, suitable especially for assistance of visually impaired people and autonomous mobile systems. This bio-inspired model relies on the interactions between formal models of three types of neurons identified in the mammals’ brain implied in navigation tasks, namely place cells, grid cells, and head direction cells, to construct a topological model of the environment under the form of a decentralized navigation graph. Previously tested in virtual environments, this model demonstrated a high tolerance to motion drift, making possible to map large environments without the need to correct it to handle such drifts, and robustness to environment changes. The presented implementation is based on a stereoscopic camera, and is evaluated on its possibilities to map and guide a person or an autonomous mobile robot in an unknown real environment. The evaluation results confirm the effectiveness of the proposed bio-inspired navigation model to build a path map, localize and guide a person through this path. The model predictions remain robust to environment changes, and allow to estimate traveled distances with an error rate below 3% over test paths, up to 100m. The tests performed on a robotic platform also demonstrated the pertinence of navigation data produced by this navigation model to guide an autonomous system. These results open the way toward efficient wearable assistive devices for visually impaired people independent navigation.

An Autonomous Mobile Combination Disinfection System.

To address the common drawbacks of current disinfection robots, which include the potential for secondary environmental pollution, disinfection dead corners, and low efficiency, in this paper, an autonomous mobile combination disinfection system is proposed. The system utilizes ultraviolet (UV) radiation and a low-concentration hydrogen peroxide aerosol to kill pathogens. It comprises three parts: a human-computer interface, a mobile robot, and disinfection equipment. A disinfection process model with continuous and fixed-point modes was established, and the effective disinfection range, speed, and duration were quantitatively calculated. The developed prototype was tested on-site by a professional third-party testing agency. The experimental results demonstrated that the combination disinfection robot achieved a 92.95% disinfection rate of natural airborne bacteria in a room measuring 22 square meters with a height of 2.8 m in just 30 min. The disinfection efficiency is at least 25% higher compared to standalone UV lamp disinfection and also exhibits a noticeable improvement over standalone hydrogen peroxide aerosol disinfection. The system enables the environmentally friendly, rapid, efficient, and all-encompassing disinfection of natural airborne bacteria. Finally, various disinfection solutions and recommendations for different application scenarios and requirements are provided.

Assessing Forest Traversability for Autonomous Mobile Systems Using Close-Range Airborne Laser Scanning

Advances in sensor technology and computing performance has brought us into an era of digital forestry where a forest environment can be digitally replicated. At the same time, an increasing interest in the use of unmanned vehicles and other autonomous mobile systems (AMSs) in forest mapping and operations has emerged. However, a forest is an unstructured and rather complex environment for AMSs to operate in, and usually some kind of a priori information of traversability is required. The aim of this study was to assess forest traversability for AMSs using high-density airborne laser scanning (ALS) point clouds. It was assumed that such point clouds acquired from a helicopter flying at a low altitude can be used to characterise vegetation obstacles affecting forest traversability. A voxel-based vegetation occupancy analysis was carried out with the aim to detect open space to define traversable three-dimensional space. The experimental setup included seven sample plots (32×32 m) representing diverse boreal forest structures. Terrestrial laser scanning (TLS) was used for obtaining reference for vegetation occupancy. Comparison between ALS and TLS revealed an overall accuracy of 0.85–0.94 with a recall of 0.78–0.91 and a precision of 0.62–0.74 for ALS-based voxel classification for vegetation occupancy depending on forest structure. This implies that up to 91% of voxels assigned a classification »occupied« based on the TLS could be correctly classified using the ALS, while up to 74% of voxels assigned a classification »occupied« using the ALS were occupied based on the TLS. Density of low vegetation accounted for 83% of the variation in accuracy and precision. The feasibility of vegetation occupancy information to be used by an AMS for navigation was also demonstrated. It was assumed that the ALS data convey as sufficient information of AMS path planning as does the TLS data. The experiments showed that out of 1393 randomly generated paths based on empty space detected by the ALS, 72% were considered feasible when validated with the TLS data. The success rate in path planning varied from 0.54 to 0.92 between the sample plots and was seemingly affected by vegetation density that accounted for 53% of variation in success rate. Altogether, the demonstrated possibility to predefine forest traversability using remote sensing will support the use of AMSs in forestry.

Analysis mobile fly area scanning systems

The rapid development of intelligent microprocessor technology, the availability of global navigation systems (GPS) and inertial imaging units (IMU) with the progress of the creation of a new generation of mobile autonomous systems from a distance This is the concept of autonomous robots and unmanned aerial vehicles. Recently, actively using unmanned aerial vehicles (UAVs) and mobile robots, they are solving the tasks of photogrammetric and laser scanning of the terrain quite quickly. High spatial resolution data collected from available platforms such as satellites and manned aircraft are typically in the range of 20–50 cm/pixel, while UAVs are capable of flying at much lower altitudes and can therefore collect images with a much higher resolution of the image.
 Mobile robots and UAVs in combination with modern technologies of digital video and photo processing, as well as programs for intelligent image recognition, have gained actual use, both for military tasks and in the civil sphere as a tool for remote sensing of territories and infrastructure high-resolution objects. On the basis of scanned UAV images, 3D models of the terrain can be reproduced.
 This article deals with the analysis of technical means of mobile scanning of territories and some well-known algorithms for controlling the scanning system.

An Extracurricular Project-Based Training Course in Innovation and Entrepreneurship Delivered to a Transdisciplinary Group of Students in Engineering, Social Sciences, Arts and Medicine

Limited information is available on the design of combined innovation and entrepreneurship training courses, and with even less available on delivering such courses to multidisciplinary teams. We designed an extracurricular project-based training course in innovation and entrepreneurship and delivered it to transdisciplinary teams of students who were then asked to build an autonomous mobile system for healthcare applications and to create a business model to implement it. The course was created through a backward educational design and design thinking to assist teams of students in transforming an idea into a technical solution and developing it into a business model through a project-based learning experience. A transdisciplinary group of 31 students, mostly in bachelor’s degree programs, worked either all together, in plenary sessions, in subject-specific groups, or in teams of 5–6 members, under the guidance of their trainers during a ten-day summer school. We used questionnaires to evaluate the changes in student perceptions and a satisfaction survey to evaluate the students’ experience. The qualitative observations together with the results of the quantitative instruments revealed positive changes in student perceptions towards innovation and entrepreneurship. Additionally, the satisfaction survey offered positive feedback and some recommendations for further improvement. The implementation of the course showed that planning is important, but flexibility is essential, that covering fewer topics might lead to deeper understanding, and that managing expectations and practicing empathy are crucial.

Deep LiDAR-Radar-Visual Fusion for Object Detection in Urban Environments

Robust environmental sensing and accurate object detection are crucial in enabling autonomous driving in urban environments. To achieve this goal, autonomous mobile systems commonly integrate multiple sensor modalities onboard, aiming to enhance accuracy and robustness. In this article, we focus on achieving accurate 2D object detection in urban autonomous driving scenarios. Considering the occlusion issues of using a single sensor from a single viewpoint, as well as the limitations of current vision-based approaches in bad weather conditions, we propose a novel multi-modal sensor fusion network called LRVFNet. This network effectively combines data from LiDAR, mmWave radar, and visual sensors through a deep multi-scale attention-based architecture. LRVFNet comprises three modules: a backbone responsible for generating distinct features from various sensor modalities, a feature fusion module utilizing the attention mechanism to fuse multi-modal features, and a pyramid module for object reasoning at different scales. By effectively fusing complementary information from multi-modal sensory data, LRVFNet enhances accuracy and robustness in 2D object detection. Extensive evaluations have been conducted on the public VOD dataset and the Flow dataset. The experimental results demonstrate the superior performance of our proposed LRVFNet compared to state-of-the-art baseline methods.

Reinforcement Learning for Dynamic Pricing of Shared-Use Autonomous Mobility Systems Considering Heterogeneous Users: Model Development and Scenario Testing

A key aspect of the success of shared-use autonomous mobility systems will be the ability to price rides in real time. As these services become more prevalent, it becomes of high importance to detect shifts in behavior to quickly optimize the system and ensure system efficiency and economic viability. Therefore, (1) pricing algorithms should be able to price rides according to complex underlying demand functions with heterogeneous customers and (2) the algorithm should be able detect nonstationary behavior (e.g., changing customers’ willingness to pay) from its previously learnt decisions and alter its pricing mechanism accordingly. We formulate a dynamic pricing and learning problem as a Markov decision process and subsequently solve it through a reinforcement learning (RL) algorithm, with heterogeneous customers accepting the trip characteristics (price, expected wait time) probabilistically. Insights from a fixed fleet operation of an autonomous private ridesourcing system in Chicago are presented. Given our formulation of the demand model, the algorithm learns in 25 days, increasing revenue by 90% and decreasing customer wait times by 90% compared to day 5. After gathering insights from the RL algorithm and applying optimal static pricing (i.e., a constant specific surge multiplier), we find that RL can achieve near 90% optimality in revenue. The RL algorithm, nevertheless, proves to be robust. Two scenarios are tested where a sudden shock occurs or customers slowly change their willingness to pay, illustrating that RL can quickly adapt its parameters to the situation.

RSS-LIWOM: Rotating Solid-State LiDAR for Robust LiDAR-Inertial-Wheel Odometry and Mapping

Solid-state LiDAR offers multiple advantages over mechanism mechanical LiDAR, including higher durability, improved coverage ratio, and lower prices. However, solid-state LiDARs typically possess a narrow field of view, making them less suitable for odometry and mapping systems, especially for mobile autonomous systems. To address this issue, we propose a novel rotating solid-state LiDAR system that incorporates a servo motor to continuously rotate the solid-state LiDAR, expanding the horizontal field of view to 360°. Additionally, we propose a multi-sensor fusion odometry and mapping algorithm for our developed sensory system that integrates an IMU, wheel encoder, motor encoder and the LiDAR into an iterated Kalman filter to obtain a robust odometry estimation. Through comprehensive experiments, we demonstrate the effectiveness of our proposed approach in both outdoor open environments and narrow indoor environments.

"OPTIMIZATION OF MOBILE SYSTEMS FOR DISABLED PEOPLES USING 3D PRINTING "

"Product innovation involves an understanding of technology through the introduction of a new product or a product that has significantly improved characteristics with the purpose of upgrading or even changing, in this case, the mobility landscape. Following the analysis of the literature in the field of methodologies for the development of an existing product or the design of a new product, the stages and activities that will be completed in order to develop an autonomous mobile system for transporting people with disabilities and transforming a manual wheelchair in an electrically operated wheelchair were established. On a basic manual wheelchair, a system will be attached to ensure overcoming the obstacles and to guarantee the ability to negociate effectively. The components, along with the control system and source energy, were mounted and tested to determine the behavior of the developed system. Various components (external and internal support, toothed crown, pinion, mobile element) were 3D printed in order to explore improving of frame attachment, the rotational movement transmission of the motor and the technical solutions for attaching the crown to the wheel. Key words: product design, wheelchair, simulation, 3D printing, optimization, mobile system. "

Cognitive Tools for Abstract Thinking Autonomous Intelligent Mobile Systems

The actual problems of artificial intelligence related to the development of tools for abstract thinking of autonomous intelligent mobile systems are being solved, which allow planning purposeful behavior in hard-to-reach and aggressive environments for humans. Cognitive tools are proposed that provide intelligent systems with the ability to organize purposeful multi-stage activities related to solving complex problems, when a behavior plan is automatically built in some conditions of a problematic environment, and a given behavior goal is achieved in other operating conditions that are beyond the resolution of technical vision. An important feature of the proposed typical elements of knowledge representation and processing is that they allow intelligent systems to organize the output of solving complex problems, relying only on the data stored in the knowledge representation model and coming from the current operating conditions. In the general case, the developed knowledge model of intelligent systems for various purposes consists of declarative and procedural typical elements of their representation. For a formal description of typical elements of declarative knowledge representation, traditional semantic networks and various sets of restrictions are used, reflecting additional conditions for the future functioning of autonomous mobile intelligent systems. As for the formal description of the typical elements of the representation of procedural knowledge, regardless of a specific subject area, fuzzy semantic networks are used for this. This allows autonomous intelligent mobile systems to adapt to specific operating conditions in underdetermined problematic environments and perform complex tasks formulated by them on this basis. The practical significance of the results obtained lies in the effectiveness of their use for the development of intelligent problem solvers that provide autonomous intelligent mobile systems for various purposes with the ability to perform complex tasks in a priori underdetermined problematic environments by adapting the purposeful activity plan formed in general form to specific current operating conditions.

BoW3D: Bag of Words for Real-Time Loop Closing in 3D LiDAR SLAM

Loop closing is a fundamental part of simultaneous localization and mapping (SLAM) for autonomous mobile systems. In the field of visual SLAM, bag of words (BoW) has achieved great success in loop closure. The BoW features for loop searching can also be used in the subsequent 6-DoF loop correction. However, for 3D LiDAR SLAM, the state-of-the-art methods may fail to effectively recognize the loop in real time, and usually cannot correct the full 6-DoF loop pose. To address this limitation, we present a novel Bag of Words for real-time loop closing in 3D LiDAR SLAM, called BoW3D. Our method not only efficiently recognizes the revisited loop places, but also corrects the full 6-DoF loop pose in real time. BoW3D builds the bag of words based on the 3D LiDAR feature LinK3D, which is efficient, pose-invariant and can be used for accurate point-to-point matching. We furthermore embed our proposed method into 3D LiDAR odometry system to evaluate loop closing performance. We test our method on public dataset, and compare it against other state-of-the-art algorithms. BoW3D shows better performance in terms of F1 max and extended precision scores on most scenarios. It is noticeable that BoW3D takes an average of 48 ms to recognize and correct the loops on KITTI 00 (includes 4K+ 64-ray LiDAR scans), when executed on a notebook with an Intel Core i7 @2.2 GHz processor. We release the implementation of our method here: https://github.com/YungeCui/BoW3D.

Optimal Clustering of Point Cloud by 2D-LiDAR Using Kalman Filter

Light detection and ranging (LiDAR) has been the primary sensor for autonomous mobility and navigation system owing to its stability. Although multiple-channel LiDAR (3D-LiDAR) can obtain dense point clouds that provide optimal performance for several tasks, the application scope is limited by its high-cost. When employing single channel LiDAR (2D-LiDAR) as a low-cost alternative, the quantity and quality of the point cloud cause conventional methods to perform poorly in clustering and tracking tasks. In particular, when handling multiple pedestrian scenarios, the point cloud is not distinguished and clustering is unable to succeed. Hence, we propose an optimized clustering method combined with a Kalman filter (KF) for simultaneous clustering and tracking applicable to 2D-LiDAR.

The Principle of Organization of Motivational Behavior and Automatic Goal Setting of Autonomous Intelligent Mobile Systems

The expediency of organizing the motivational behavior of autonomous intelligent mobile systems focused on solving various complex problems in unstable a priori undescribed problematic environments is substantiated. That need is due to the fact that this type of goal-seeking behavior allows intelligent systems of various purposes to ensure safe and efficient activity in unstable operating conditions. A model of knowledge representation of autonomous intelligent mobile systems is proposed without regard to a specific subject area and built on the basis of active fuzzy semantic networks. In such networks, vertices are labeled with slots that have many characteristics, which enables in the process of goal-seeking activity to label active vertices with objects and events occurring in the problematic environment that meet their requirements. Edges in such networks are labeled with generalized fuzzy values of corresponding conditions that must be met in a problematic environment between an autonomous intelligent mobile system, various objects, and occurring events. This model of a formalized description of various situations and subsituations of the problematic environment allows intelligent systems to adapt to a priori undescribed operating conditions and, on this basis, automatically plan goal-seeking activities under conditions of uncertainty. To control the motivational behavior and self-organization of autonomous intelligent mobile systems, tools have been developed intended to identify threats to productive activities that arise in a problematic environment. A generalized production-based model of knowledge representation has been built without regard to a specific subject area, which allows intelligent systems to quickly respond to various types of security threats that arise in a problematic environment and maintain operability in the process of performing complex tasks. In conclusion, one of the effective ways of further development of the proposed principle of organizing the safe and efficient operation of autonomous intelligent mobile systems is outlined, which is related to the management of their collective activities in the process of performing a complex task with changes occurring spontaneously in a problem environment, accompanied by the emergence of various types of threats in it that prevent their effective operation.

Relocalization based on millimeter wave radar point cloud for visually degraded environments

AbstractSimultaneously localization and mapping (SLAM) has been widely used in autonomous mobile systems to fulfill autonomous navigation. Relocalization plays an important role in SLAM for closing the loop and eliminating the drift of pose estimation. Traditional methods mostly rely on LiDAR or camera sensors, which may degrade or even fail in rainy or dusty situations or with large illumination changes. In this article, we explore the use of low‐cost commercial millimeter wave (mmWave) radars and propose a noval mmWave radar point cloud‐based relocalization method. Our method first pre‐processes the radar point cloud and, based on that, achieves fast 3‐DOF pose estimation for the robot. We build a prototype and thoroughly evaluate our method using data sets collected by our platform in four complex environments, including street, park, road, and water surface scenarios. The experimental results show that our method consistently outperforms other baseline methods including the vision‐based counterparts, especially in the visual degraded scenes.

Envisioning a Shared Low-Speed Light-Weight Autonomous Mobility (SLLAM) System on Continuous Pedestrian Access Routes: A Case Study in Tuscaloosa, Alabama

Envisioning a Shared Low-Speed Light-Weight Autonomous Mobility (SLLAM) System on Continuous Pedestrian Access Routes: A Case Study in Tuscaloosa, Alabama

DESIGN OF AUTONOMOUS ROBOTIC PLATFORMS BASED ON MODEL DATA

The article presents a method for selecting the configuration of the shooting equipment of a robotic platform, associated with the use of virtual digital models of sensors and a scene. The topic is really actual, since a number of complex technical and interesting scientific problems of computer vision are currently associated with the use of autonomous mobile systems. Technical difficulties are determined, first of all, by the fact that such complexes are equipped with many different sensors, the data from which must be processed in a single coordinate system with high accuracy. To do this, it is necessary to calibrate the system using an object with a special markup, the use of which allows you to effectively use computer vision algorithms and make the process fully automatic. The accuracy of the relative orientation of the sensors, which can be achieved in this case, can be considered a quality criterion when choosing survey equipment. In the process of designing a platform, it is rather difficult to determine in advance how many sensors and their placement is optimal in terms of the result - obtaining a high quality digital scene model. In this situation, it is natural to compare different hardware configurations and choose from them. However, to obtain the required amount of real data, a significant amount of time is required to work with complex expensive equipment. The key to solving the problem may be the use of virtual data - a realistic digital model of the sensors and the scene, synthesized using the graphics engine. This paper presents the process of designing a promising vehicle, including the selection and placement of shooting equipment on the platform – cameras and lidars. The best configuration is selected from the three proposed according to the criterion of the accuracy of restoring the positions of the sensors during the calibration process. The application software was developed on the basis of the Unreal Engine 4 graphics engine. Digital models of the vehicle, sensors and scene were created in accordance with the parameters of the real prototype. Taking into account the technical characteristics of the sensors, three options for their placement have been created. For each variant, s