Autonomous exploration of complex, unknown environments is a cutting-edge task not entirely solved by the scientific community. When an agent needs to explore a maze without any a priori information about the environment, the lack of proper destinations and explicit task objectives make traditional navigation policies inappropriate. While the literature presents some sporadic deterministic systems able to face the tasks, learning approaches still need an adequate investigation which could prove them to be more suitable and versatile for this purpose. In this paper, we present MARS, a path planner that exploits swarms of robots to optimize the exploration of complex unknown environments, such as mazes. To make the solution scalable, the proposed method exploits two cooperating modules: local and global planners. The local planner is modeled as a Markov Decision Process (MDP) and trained as a Reinforcement Learning (RL) multi-agent system. Each agent has access to image representations of a section of the global map, always centered in the robot reference frame, and decides the next navigation goal to complete the local exploration. The global planner is a deterministic system that recovers the navigation when a local solution is unavailable. The robots share the explored section with peers when they meet in a rendez-vous. We compared our approach to a single deterministic agent, a single RL agent and a close-to-optimal deterministic approach which deploys five greedy agents. The simulation results demonstrate MARS' efficiency, reaching near-optimal levels in significantly less time.

Satellites are used for a multitude of applications, including communications, Earth observation, and space science. Neural networks and deep learning-based approaches now represent the state-of-the-art to enhance the performance and efficiency of these tasks. Given that satellites are susceptible to various faults, one critical application of Artificial Intelligence (AI) is fault detection. However, despite the advantages of neural networks, these systems are vulnerable to radiation errors, which can significantly impact their reliability. Ensuring the dependability of these solutions requires extensive testing and validation, particularly using fault injection methods. This study analyses a physics-informed (PI) real-valued non-volume preserving (Real NVP) normalizing flow model for fault detection in space systems, with a focus on resilience to Single-Event Upsets (SEUs). We present a customized fault injection framework in TensorFlow to assess neural network resilience. Fault injections are applied through two primary methods: Layer State injection, targeting internal network components such as weights and biases, and Layer Output injection, which modifies layer outputs across various activations. Fault types include zeros, random values, and bit-flip operations, applied at varying levels and across different network layers. Our findings reveal several critical insights, such as the significance of bit-flip errors in critical bits, that can lead to substantial performance degradation or even system failure. With this work, we aim to exhaustively study the resilience of Real NVP models against errors due to radiation, providing a means to guide the implementation of fault tolerance measures.

Achieving social acceptance is one of the main goals of Social Robotic Navigation. Despite this topic has received increasing interest in recent years, most of the research has focused on driving the robotic agent along obstacle-free trajectories, planning around estimates of future human motion to respect personal distances and optimize navigation. However, social interactions in everyday life are also dictated by norms that do not strictly depend on movement, such as when standing at the end of a queue rather than cutting it. In this paper, we propose a novel method to recognize common social scenarios and modify a traditional planner’s cost function to adapt to them. This solution enables the robot to carry out different social navigation behaviors that would not arise otherwise, maintaining the robustness of traditional navigation. Our approach allows the robot to learn different social norms with a single learned model, rather than having different modules for each task. As a proof of concept, we consider the tasks of queuing and respect interaction spaces of groups of people talking to one another, but the method can be extended to other human activities that do not involve motion.

Human-aware navigation is a complex task for mobile robots, requiring an autonomous navigation system capable of achieving efficient path planning together with socially compliant behaviors. Social planners usually add costs or constraints to the objective function, leading to intricate tuning processes or tailoring the solution to the specific social scenario. Machine Learning can enhance planners’ versatility and help them learn complex social behaviors from data. This work proposes an adaptive social planner, using a Deep Reinforcement Learning agent to dynamically adjust the weighting parameters of the cost function used to evaluate trajectories. The resulting planner combines the robustness of the classic Dynamic Window Approach, integrated with a social cost based on the Social Force Model, and the flexibility of learning methods to boost the overall performance on social navigation tasks. Our extensive experimentation on different environments demonstrates the general advantage of the proposed method over static cost planners.

Single-Image Super-Resolution can support robotic tasks in environments where a reliable visual stream is required to monitor the mission, handle teleoperation or study relevant visual details. In this work, we propose an efficient Generative Adversarial Network model for real-time Super-Resolution, called EdgeSRGAN1. We adopt a tailored architecture of the original SRGAN and model quantization to boost the execution on CPU and Edge TPU devices, achieving up to 200 fps inference. We further optimize our model by distilling its knowledge to a smaller version of the network and obtain remarkable improvements compared to the standard training approach. Our experiments show that our fast and lightweight model preserves considerably satisfying image quality compared to heavier state-of-the-art models. Finally, we conduct experiments on image transmission with bandwidth degradation to highlight the advantages of the proposed system for mobile robotic applications.

Indoor autonomous navigation requires a precise and accurate localization system able to guide robots through cluttered, unstructured and dynamic environments. Ultra-wideband (UWB) technology, as an indoor positioning system, offers precise localization and tracking, but moving obstacles and non-line-of-sight occurrences can generate noisy and unreliable signals. That, combined with sensors noise, unmodeled dynamics and environment changes can result in a failure of the guidance algorithm of the robot. We demonstrate how a power-efficient and low computational cost point-to-point local planner, learnt with deep reinforcement learning (RL), combined with UWB localization technology can constitute a robust and resilient to noise short-range guidance system complete solution. We trained the RL agent on a simulated environment that encapsulates the robot dynamics and task constraints and then, we tested the learnt point-to-point navigation policies in a real setting with more than two-hundred experimental evaluations using UWB localization. Our results show that the computational efficient end-to-end policy learnt in plain simulation, that directly maps low-range sensors signals to robot controls, deployed in combination with ultra-wideband noisy localization in a real environment, can provide a robust, scalable and at-the-edge low-cost navigation system solution.

Precision agriculture is a fast-growing field that aims at introducing affordable and effective automation into agricultural processes. Nowadays, algorithmic solutions for navigation in vineyards require expensive sensors and high computational workloads that preclude large-scale applicability of autonomous robotic platforms in real business case scenarios. From this perspective, our novel proposed control leverages the latest advancement in machine perception and edge AI techniques to achieve highly affordable and reliable navigation inside vineyard rows with low computational and power consumption. Indeed, using a custom-trained segmentation network and a low-range RGB-D camera, we are able to take advantage of the semantic information of the environment to produce smooth trajectories and stable control in different vineyards scenarios. Moreover, the segmentation maps generated by the control algorithm itself could be directly exploited as filters for a vegetative assessment of the crop status. Extensive experimentations and evaluations against real-world data and simulated environments demonstrated the effectiveness and intrinsic robustness of our methodology.

A hybrid Maximum Power Point Tracking (MPPT) method without using an irradiance sensor is proposed in this paper. The hybrid MPPT method is a combination of conventional Perturb and Observe (P&O) and Fractional Short Circuit Current (FSCC) MPPT technique. The proposed hybrid MPPT decides intelligently about the initial operating point of P&O by using the FSCC MPPT method. This method requires no sensor for irradiation measurement because it detects intelligently about the change in irradiance. Therefore, under dynamic weather conditions the decision about measuring the initial operating point of P&O is intelligent. After finding the initial operating point the system shifts to the conventional P&O method and starts to perturb with a small perturbation size. The use of small perturbation steps enables the system to work with less power oscillations around the Maximum Power Point (MPP), while the use of FSCC helps in rapid tracking of MPP especially under the varying environmental conditions. Thus, the proposed method is fast in tracking the MPP with less power oscillations and hence better performance in terms of energy harvesting when compared with the conventional P&O technique. Impedance matching between solar panel and load is achieved using a DC-DC boost converter and the proposed method is simulated in MATLAB/SIMULINK environment. The system is simulated against the steady and dynamic weather conditions. The results shows that under steady weather condition the proposed method harvests 5% extra energy as compared with the conventional P&O. For dynamic weather conditions the proposed method harness an additional 3.5% energy when compared with the P&O method, which makes it very promising.

We propose a semi-automatic HW/SW codesign flow for low-power and low-cost Neuro-Fuzzy embedded systems. Applications range from fast prototyping of embedded systems to high-speed simulation of Simulink models and rapid design of Neuro-Fuzzy devices. The proposed codesign flow works with different technologies and architectures (namely, software, digital and analog). We have used The Mathworks' Simulink environment for functional specification and for analysis of performance criteria such as timing (latency and throughput), power dissipation, size and cost. The proposed flow can exploit trade-offs between SW and HW as well as between digital and analog implementations, and it can generate, respectively, the C, VHDL and SKILL codes of the selected architectures.

Artificial neural networks (NNs) and fuzzy systems are highly parallel structures that consist of a large number of fully interconnected elementary nonlinear units (called neurons). We present a digital hardware implementation of NNs combined with a host computer (DSP systems, PCs, etc.) resulting in a powerful system used in different applications. The NNs process most of the tasks, while the host performs signal pre-processing and learning algorithms. In some control applications the host also tracks some discrete states of the plant by implementing a finite state automata and/or verifying plant safety boundaries operations. The tight link with the host allows the NN hardware to be very simple since several operations related with the NNs (learning, weights initialization, etc.) can be performed by the host computer. The board can be used to implement intelligent control paradigms mixing neuro-fuzzy algorithms with finite state automata and/or digital control algorithms.

This paper describes an existing silicon implementation of an artificial neural system based on coherent pulse width and edge modulation techniques. A chip set with different neural functions has been conceived, manufactured and tested. Neural circuits have been optimized for lowest computation energy and highest reconfigurability. The main device is a 32 x 32 synaptic array consuming 10 mW of power at 140 MCPS. Synapsis size is about 7.200 microns 2 using a standard 1.5 microns CMOS technology. The problem of interfacing robotic sensors and actuators is addressed: voltage, current and resistance-based sensors are considered for the measurement of physical quantities such as temperature, pressure, strain, etc. Low resolution imaging sensors for robotic vision are also considered.