Navigation Tasks Research Articles

Effects of Feedback Type on Touchscreen Navigation for Younger and Older Users

Abstract This study investigates how feedback from different modalities affect window navigation tasks performed on a touchscreen by younger and older users. Thirty-two younger and 32 older users moved a window toward one of four or eight target areas with no feedback or visual, tactile or visual–tactile feedback. Visual feedback produced significantly higher accuracy and shorter task completion times than tactile feedback. Generally, visual–tactile feedback did not produce statistically higher accuracy or shorter task completion time than visual feedback. However, visual–tactile feedback prevented both younger and older users from overshooting the windows. Additionally, when users attempted to move the window to one of the eight target areas, the visual–tactile feedback helped younger users perform better. According to theories about human processing, the reason for the subtle advantage of visual–tactile feedback is that the unintuitive tactile force induces a higher cognitive load than straightforward visual information during window navigation. Nevertheless, visual–tactile feedback elicited the best user experience, as measured by questionnaires on emotion and fatigue, gesture usability and preference. Furthermore, more older users prefer visual–tactile feedback than younger users. Therefore, visual–tactile feedback should be considered, and the tactile feedback should be modified to be more intuitive in window navigation tasks. Research Highlights Visual feedback improved task performance more than tactile feedback in window navigation task. Visual–tactile feedback did not improve task performance compared to visual feedback. Visual–tactile feedback prevented users from overshooting the window. More older users than younger users prefer visual–tactile feedback.

A recurrent network model of planning explains hippocampal replay and human behavior

When faced with a novel situation, people often spend substantial periods of time contemplating possible futures. For such planning to be rational, the benefits to behavior must compensate for the time spent thinking. Here, we capture these features of behavior by developing a neural network model where planning itself is controlled by the prefrontal cortex. This model consists of a meta-reinforcement learning agent augmented with the ability to plan by sampling imagined action sequences from its own policy, which we call ‘rollouts’. In a spatial navigation task, the agent learns to plan when it is beneficial, which provides a normative explanation for empirical variability in human thinking times. Additionally, the patterns of policy rollouts used by the artificial agent closely resemble patterns of rodent hippocampal replays. Our work provides a theory of how the brain could implement planning through prefrontal–hippocampal interactions, where hippocampal replays are triggered by—and adaptively affect—prefrontal dynamics.

Designing Reward Functions Using Active Preference Learning for Reinforcement Learning in Autonomous Driving Navigation

This study presents a method based on active preference learning to overcome the challenges of designing reward functions for autonomous navigation. Results obtained from training with artificially designed reward functions may not accurately reflect human intentions. We focus on the limitations of traditional reward functions, which often fail to facilitate complex tasks in continuous state spaces. We propose the adoption of active preference learning to resolve these issues and to generate reward functions that align with human preferences. This approach leverages an individual’s subjective preferences to guide an agent’s learning process, enabling the creation of reward functions that reflect human desires. We utilize mutual information to generate informative queries and apply information gained to balance the agent’s uncertainty with the human’s response capacity, encouraging the agent to pose straightforward and informative questions. We further employ the No-U-Turn Sampler (NUTS) method to refine the belief model, which outperforms that constructed using the Metropolis algorithm. Subsequently, we retrain the agent using reward weights derived from active preference learning. As a result, our autonomous driving vehicle can navigate between random starting and ending points without dependence on high-precision maps or routing, relying solely on its forward vision. We validate our approach’s performance within the CARLA simulation environment. Our algorithm significantly improved the success rate of autonomous driving navigation tasks that originally failed due to artificially designed rewards, increasing it to approximately 60%. Experimental results show significant improvement over the baseline algorithm, providing a solid foundation for enhancing navigation capabilities in autonomous driving systems and advancing the field of autonomous driving intelligence.

A Local-and-Global Attention Reinforcement Learning Algorithm for Multiagent Cooperative Navigation.

The cooperative navigation algorithm is the crucial technology for multirobot systems to accomplish autonomous collaborative operations, and it is still a challenge for researchers. In this work, we propose a new multiagent reinforcement learning algorithm called multiagent local-and-global attention actor-critic (MLGA2C) for multiagent cooperative navigation. Inspired by the attention mechanism, we design the local-and-global attention module to dynamically extract and encode critical environmental features. Meanwhile, based on the centralized training and decentralized execution (CTDE) paradigm, we extend a new actor-critic method to handle feature encoding and make navigation decisions. We also evaluate the proposed algorithm in two cooperative navigation scenarios: static target navigation and dynamic pedestrian target tracking. The multiple experimental results show that our algorithm performs well in cooperative navigation tasks with increasing agents.

SayNav: Grounding Large Language Models for Dynamic Planning to Navigation in New Environments

Semantic reasoning and dynamic planning capabilities are crucial for an autonomous agent to perform complex navigation tasks in unknown environments. It requires a large amount of common-sense knowledge, that humans possess, to succeed in these tasks. We present SayNav, a new approach that leverages human knowledge from Large Language Models (LLMs) for efficient generalization to complex navigation tasks in unknown large-scale environments. SayNav uses a novel grounding mechanism, that incrementally builds a 3D scene graph of the explored environment as inputs to LLMs, for generating feasible and contextually appropriate high-level plans for navigation. The LLM-generated plan is then executed by a pre-trained low-level planner, that treats each planned step as a short-distance point-goal navigation sub-task. SayNav dynamically generates step-by-step instructions during navigation and continuously refines future steps based on newly perceived information. We evaluate SayNav on multi-object navigation (MultiON) task, that requires the agent to utilize a massive amount of human knowledge to efficiently search multiple different objects in an unknown environment. We also introduce a benchmark dataset for MultiON task employing ProcTHOR framework that provides large photo-realistic indoor environments with variety of objects. SayNav achieves state-of-the-art results and even outperforms an oracle based baseline with strong ground-truth assumptions by more than 8% in terms of success rate, highlighting its ability to generate dynamic plans for successfully locating objects in large-scale new environments. The code, benchmark dataset and demonstration videos are accessible at https://www.sri.com/ics/computer-vision/saynav.

Bodily and Visual-Cognitive Navigation Aids to Enhance Spatial Recall in Mild Cognitive Impairment.

Individuals with mild cognitive impairment (MCI) syndrome often report navigation difficulties, accompanied by impairments in egocentric and allocentric spatial memory. However, studies have shown that both bodily (e.g., motor commands, proprioception, vestibular information) and visual-cognitive (e.g., maps, directional arrows, attentional markers) cues can support spatial memory in MCI. We aimed to assess navigation cues for innovative spatial training in aging. Fifteen MCI patients were recruited for this study. Their egocentric and allocentric memory recall performances were tested through a navigation task with five different virtual reality (VR) assistive encoding navigation procedures (bodily, vision only, interactive allocentric map, reduced executive load, free navigation without cues). Bodily condition consisted of an immersive VR setup to engage self-motion cues, vision only condition consisted of passive navigation without interaction, in the interactive allocentric map condition patients could use a bird-view map, in the reduced executive load condition directional cues and attentional markers were employed, and during free navigation no aid was implemented. Bodily condition improved spatial memory compared to vision only and free navigation without cues. In addition, the interactive allocentric map was superior to the free navigation without cues. Surprisingly, the reduced executive load was comparable to vison only condition. Moreover, a detrimental impact of free navigation was observed on allocentric memory across testing trials. These findings challenge the notion of an amodal representation of space in aging, suggesting that spatial maps can be influenced by the modality in which the environment was originally encoded.

Hierarchical reinforcement learning for handling sparse rewards in multi-goal navigation

Reinforcement learning (RL) has achieved remarkable advancements in navigation tasks in recent years. However, tackling multi-goal navigation tasks with sparse rewards remains a complex and challenging problem due to the long-sequence decision-making involved. Such multi-goal navigation tasks inherently incorporate a hybrid action space, where the robot needs to select a navigation endpoint first before executing primitive actions. To address the problem of multi-goal navigation with sparse rewards, we introduce a novel hierarchical RL framework named Hierarchical RL with Multi-Goal (HRL-MG). The main idea of HRL-MG is to divide and conquer the hybrid action space, splitting long-sequence decisions into short-sequence decisions. The HRL-MG framework is composed of two main modules: a selector and an actuator. The selector employs a temporal abstraction hierarchical architecture designed to specify a desired end goal based on the discrete action space. Conversely, the actuator utilizes a continuous goal-oriented hierarchical architecture developed to enact continuous action sequences to reach the desired end goal specified by the selector. In addition, we incorporate a dynamic goal detection mechanism, grounded in hindsight experience replay, to mitigate the challenges posed by sparse reward landscapes. We validated the algorithm’s efficacy on both the discrete environment Maze_2D and the continuous robotic environment MuJoCo ‘Ant’. The results indicate that HRL-MG significantly outperforms other methods in multi-goal navigation tasks with sparse rewards.

A Comparison Between A* and RRT Algorithm in Path Planning for Mobile Robot

The ability of mobile devices to navigate environments is vital. Navigation plays a critical role in averting collisions and hazardous situations.The design of paths is an essential component of robot navigation. comprising the ability to determine the optimal route from the robot's current location to a given destination. And in all state-of-the-art algorithms for robots to design the optimal route, A* and RRT algorithm stand out as the two most widely used path planning techniques, one is graph-based approach while the other is sample-based. In this paper, some variants or improvement of the two methods will be described. And a comparison about A* and RRT will be presented with appropriate criteria, which are path length and computational time. The main finding is that the A* method outperforms the Rapidly Exploring Random Tree (RRT) approach in terms of computational efficiency and path distance optimization. But it requires to consider more thoughts in applications. The comparison of the A* and RRT algorithms helps comprehend their applicability in various application circumstances. In this work, Evaluating these aspects altogether comprehensively provides a deeper understanding of the suitability of each algorithm for specific robotic navigation tasks.

Indoor navigation map design based on spatial complexity

ABSTRACT In the face of the complex interior structure of a building, indoor navigation is an important tool for people to arrive at their destination in large public buildings. In this research, we first analyze the complexity of indoor space, which mainly includes spatial entities with rich semantics, but corridors lacking in semantics, poor visibility, and multi-dimensional floors. Then, we designed various indoor navigation maps based on spatial complexity and conducted a comparative analysis between the existing maps and the ones we designed. Participants were recruited to perform three navigation tasks in the study area on-site. The results demonstrate the advantages of using a hierarchical landmark representation for improved spatial knowledge acquisition and the effectiveness of scene-enhanced directional guidance in areas with poor visibility. It is worth noting that the introduction of three-dimensional maps was found to have a positive impact on users’ spatial perception and their ability to sketch routes accurately. In summary, this evidence confirms that maps based on spatial complexity not only enhance navigation efficiency but also have the potential to improve the overall indoor navigation experience.

An Improved Dyna-Q Algorithm Inspired by the Forward Prediction Mechanism in the Rat Brain for Mobile Robot Path Planning.

The traditional Model-Based Reinforcement Learning (MBRL) algorithm has high computational cost, poor convergence, and poor performance in robot spatial cognition and navigation tasks, and it cannot fully explain the ability of animals to quickly adapt to environmental changes and learn a variety of complex tasks. Studies have shown that vicarious trial and error (VTE) and the hippocampus forward prediction mechanism in rats and other mammals can be used as key components of action selection in MBRL to support "goal-oriented" behavior. Therefore, we propose an improved Dyna-Q algorithm inspired by the forward prediction mechanism of the hippocampus to solve the above problems and tackle the exploration-exploitation dilemma of Reinforcement Learning (RL). This algorithm alternately presents the potential path in the future for mobile robots and dynamically adjusts the sweep length according to the decision certainty, so as to determine action selection. We test the performance of the algorithm in a two-dimensional maze environment with static and dynamic obstacles, respectively. Compared with classic RL algorithms like State-Action-Reward-State-Action (SARSA) and Dyna-Q, the algorithm can speed up spatial cognition and improve the global search ability of path planning. In addition, our method reflects key features of how the brain organizes MBRL to effectively solve difficult tasks such as navigation, and it provides a new idea for spatial cognitive tasks from a biological perspective.

APEX: affordance-based plan executor for indoor robotic navigation

ABSTRACT The concept of affordance is commonly described as the set of possibilities an environment offers an animal to select and execute actions. In the context of autonomous robot navigation, determining which affordances are available to a mobile agent provides crucial information to perform navigation tasks such as executing navigation plans or exploring new environments. In this work, we provide a robot with the capability of detecting changes in the set of navigation affordances that it has available as it moves through indoor spaces. Specifically, given a navigation plan represented as a sequence of high-level behaviors such as ‘turn left’ or ‘follow a corridor’, we propose the Affordance-Based Plan Executor (APEX) as a new learning-based method to execute this plan based on the identification of navigation affordances. As a relevant fact, the proposed method does not require a map of the environment at execution time. To the best of our knowledge, our work is one of the first to explicitly consider the identification of navigation affordances as a central element to implement the execution of a behavior-based navigation plan in an indoor environment. Our experiments using the Gibson simulator and environments from the Stanford 2D-3D-S dataset suggest that our approach performs well in previously seen environments in comparison to baselines, exceeding alternative approaches' success rate by up to 43%. Additionally, our approach shows promise at generalizing to previously unseen environments, exceeding the success rate of alternative approaches by up to 26.3%. Finally, our experiments confirm that explicit reasoning about affordances is key to APEX's performance.

Crater Triangle Matching Algorithm Based on Fused Geometric and Regional Features

Craters are regarded as significant navigation landmarks during the descent and landing process in small body exploration missions for their universality. Recognizing and matching craters is a crucial prerequisite for visual and LIDAR-based navigation tasks. Compared to traditional algorithms, deep learning-based crater detection algorithms can achieve a higher recognition rate. However, matching crater detection results under various image transformations still poses challenges. To address the problem, a composite feature-matching algorithm that combines geometric descriptors and region descriptors (extracting normalized region pixel gradient features as feature vectors) is proposed. First, the geometric configuration map is constructed based on the crater detection results. Then, geometric descriptors and region descriptors are established within each feature primitive of the map. Subsequently, taking the salience of geometric features into consideration, composite feature descriptors with scale, rotation, and illumination invariance are generated through fusion geometric and region descriptors. Finally, descriptor matching is accomplished by computing the relative distances between descriptors and adhering to the nearest neighbor principle. Experimental results show that the composite feature descriptor proposed in this paper has better matching performance than only using shape descriptors or region descriptors, and can achieve a more than 90% correct matching rate, which can provide technical support for the small body visual navigation task.

Design and Implementation of a Low-Cost Intelligent Unmanned Surface Vehicle

This article describes the design and construction journey of a self-developed unmanned surface vehicle (USV). In order to increase the accessibility and lower the barrier of entry we propose a low-cost (under EUR 1000) approach to the vessel construction with great adaptability and customizability. This design prioritizes minimal power consumption as a key objective. It focuses on elucidating the intricacies of both the design and assembly processes involved in creating an economical USV. Utilizing easily accessible components, the boat outlined in this study has been already participated in the 1st Aegean Ro-boat Race 2023 competition and is tailored for entry into similar robotic competitions. Its primary functionalities encompass autonomous sea navigation coupled with sophisticated collision avoidance capabilities. Finally, we studied reinforcement learning strategies for constructing a robust intelligent controller for the task of USV navigation under disturbances and we show some preliminary simulation results we have obtained.

Credit Assignment: Challenges and Opportunities in Developing Human-like Learning Agents

Temporal credit assignment is the process of distributing delayed outcomes to each action in a sequence, which is essential for learning to adapt and make decisions in dynamic environments. While computational methods in reinforcement learning, such as temporal difference (TD), have shown success in tackling this issue, it remains unclear whether these mechanisms accurately reflect how humans handle feedback delays. Furthermore, cognitive science research has not fully explored the credit assignment problem in humans and cognitive models. Our study uses a cognitive model based on Instance-Based Learning Theory (IBLT) to investigate various credit assignment mechanisms, including equal credit, exponential credit, and TD credit, using the IBL decision mechanism in a goal-seeking navigation task with feedback delays and varying levels of decision complexity. We compare the performance and process measures of the different models with human decision-making in two experiments. Our findings indicate that the human learning process cannot be fully explained by any of the mechanisms. We also observe that decision complexity affects human behavior but not model behavior. By examining the similarities and differences between human and model behavior, we summarize the challenges and opportunities for developing learning agents that emulate human decisions in dynamic environments.

Behavioral Disorders of Spatial Cognition in Patients with Mild Cognitive Impairment Due to Alzheimer's Disease (The BDSC-MCI Project): Ecological Validity of the Corsi Learning Suvra-Span Test.

Spatial navigation deficits are reported as early symptoms of Alzheimer's disease (AD) alongside episodic memory ones. The aim of the present study was to ascertain whether neuropsychological deficits of visuospatial long-term memory can predict behavioral alterations during the navigation of older adults in novel urban environments along the normal aging-dementia continuum of the Alzheimer's type. A total of 24 community-dwelling patients with Mild Cognitive Impairment (MCI) due to AD, 27 individuals with subjective cognitive decline (SCD), and 21 healthy controls were assessed in terms of their sequential egocentric and allocentric navigation abilities by using a modified version of the Detour Navigation Test, and neuropsychologically tested by the Corsi learning suvra-span (CLSS) test. Generalized linear models were adopted to verify whether the scores obtained by the three groups in the CLSS test predicted wrong turns and moments of hesitation during the navigation task, with the results presented as topographical disorientation scores. Higher scores in the CLSS test predicted fewer wrong turns (b = -0.05; z = -2.91; p = 0.004; net of between-groups differences) and moments of hesitation for patients with MCI due to AD (b = -0.14; z = -2.43; p = 0.015), and individuals with SCD (b = -0.17; z = -3.85; p < 0.001). Since the CLSS test has been reported to be a reliable measure of ecological navigational abilities in the progression towards AD dementia, we recommend its use in clinical practice and highlight implications for future research.

A non-canonical visual cortical-entorhinal pathway contributes to spatial navigation

Visual information is important for accurate spatial coding and memory-guided navigation. As a crucial area for spatial cognition, the medial entorhinal cortex (MEC) harbors diverse spatially tuned cells and functions as the major gateway relaying sensory inputs to the hippocampus containing place cells. However, how visual information enters the MEC has not been fully understood. Here, we identify a pathway originating in the secondary visual cortex (V2) and directly targeting MEC layer 5a (L5a). L5a neurons served as a network hub for visual processing in the MEC by routing visual inputs from multiple V2 areas to other local neurons and hippocampal CA1. Interrupting this pathway severely impaired visual stimulus-evoked neural activity in the MEC and performance of mice in navigation tasks. These observations reveal a visual cortical-entorhinal pathway highlighting the role of MEC L5a in sensory information transmission, a function typically attributed to MEC superficial layers before.

Self-Supervised 3-D Semantic Representation Learning for Vision-and-Language Navigation.

In vision-and-language navigation (VLN) tasks, most current methods primarily utilize RGB images, overlooking the rich 3-D semantic data inherent to environments. To rectify this, we introduce a novel VLN framework that integrates 3-D semantic information into the navigation process. Our approach features a self-supervised training scheme that incorporates voxel-level 3-D semantic reconstruction to create a detailed 3-D semantic representation. A key component of this framework is a pretext task focused on region queries, which determines the presence of objects in specific 3-D areas. Following this, we devise an long short-term memory (LSTM)-based navigation model that is trained using our 3-D semantic representations. To maximize the utility of these 3-D semantic representations, we implement a cross-modal distillation strategy. This strategy encourages the RGB model's outputs to emulate those from the 3-D semantic feature network, enabling the concurrent training of both branches to merge RGB and 3-D semantic data effectively. Comprehensive evaluations on both the R2R and R4R datasets reveal that our method significantly enhances performance in VLN tasks.

Smart Delivery Assignment through Machine Learning and the Hungarian Algorithm

Intelligent transportation and advanced mobility techniques focus on helping operators to efficiently manage navigation tasks in smart cities, enhancing cost efficiency, increasing security, and reducing costs. Although this field has seen significant advances in developing large-scale monitoring of smart cities, several challenges persist concerning the practical assignment of delivery personnel to customer orders. To address this issue, we propose an architecture to optimize the task assignment problem for delivery personnel. We propose the use of different cost functions obtained with deterministic and machine learning techniques. In particular, we compared the performance of linear and polynomial regression methods to construct different cost functions represented by matrices with orders and delivery people information. Then, we applied the Hungarian optimization algorithm to solve the assignment problem, which optimally assigns delivery personnel and orders. The results demonstrate that when used to estimate distance information, linear regression can reduce estimation errors by up to 568.52 km (1.51%) for our dataset compared to other methods. In contrast, polynomial regression proves effective in constructing a superior cost function based on time information, reducing estimation errors by up to 17,143.41 min (11.59%) compared to alternative methods. The proposed approach aims to enhance delivery personnel allocation within the delivery sector, thereby optimizing the efficiency of this process.

Evidence of elevated situational awareness for active duty soldiers during navigation of a virtual environment.

U.S. service members maintain constant situational awareness (SA) due to training and experience operating in dynamic and complex environments. Work examining how military experience impacts SA during visual search of a complex naturalistic environment, is limited. Here, we compare Active Duty service members and Civilians' physiological behavior during a navigational visual search task in an open-world virtual environment (VE) while cognitive load was manipulated. We measured eye-tracking and electroencephalogram (EEG) outcomes from Active Duty (N = 21) and Civilians (N = 15) while they navigated a desktop VE at a self-regulated pace. Participants searched and counted targets (N = 15) presented among distractors, while cognitive load was manipulated with an auditory Math Task. Results showed Active Duty participants reported significantly greater/closer to the correct number of targets compared to Civilians. Overall, Active Duty participants scanned the VE with faster peak saccade velocities and greater average saccade magnitudes compared to Civilians. Convolutional Neural Network (CNN) response (EEG P-300) was significantly weighted more to initial fixations for the Active Duty group, showing reduced attentional resources on object refixations compared to Civilians. There were no group differences in fixation outcomes or overall CNN response when comparing targets versus distractor objects. When cognitive load was manipulated, only Civilians significantly decreased their average dwell time on each object and the Active Duty group had significantly fewer numbers of correct answers on the Math Task. Overall, the Active Duty group explored the VE with increased scanning speed and distance and reduced cognitive re-processing on objects, employing a different, perhaps expert, visual search strategy indicative of increased SA. The Active Duty group maintained SA in the main visual search task and did not appear to shift focus to the secondary Math Task. Future work could compare how a stress inducing environment impacts these groups' physiological or cognitive markers and performance for these groups.

Validating Synthetic Data for Perception in Autonomous Airport Navigation Tasks

Autonomous navigation within airport environments presents significant challenges, mostly due to the scarcity of accessible and labeled data for training autonomous systems. This study introduces an innovative approach to assess the performance of vision-based models trained on synthetic datasets, with the goal of determining whether simulated data can train and validate navigation operations in complex airport environments. The methodology includes a comparative analysis employing image processing techniques and object detection algorithms. A comparative analysis of two different datasets was conducted: a synthetic dataset that mirrors real airport scenarios, generated using the Microsoft Flight Simulator 2020®video game, and a real-world dataset. The results indicate that models trained on a combination of both real and synthetic images perform much better in terms of adaptability and accuracy compared to those trained only on one type of dataset. This analysis makes a significant contribution to the field of autonomous airport navigation and offers a cost-effective and practical solution to overcome the challenges of dataset acquisition and algorithm validation. It is thus believed that this study lays the groundwork for future advancements in the field.