Journal of Intelligent & Robotic Systems最新文献

This paper proposes a new 3D Lyapunov guidance vector field(3D-LGV) avoidance strategy based on reinforcement learning for the satellite evasion and interception problem. Combining it with the interfered fluid dynamical system (IFDS) enables the satellite to evade and smoothly enter orbit according to the state of the intercepting satellite in real time. 3D-LGV provides an initial flow field approaching an elliptical orbit, while IFDS provides a perturbed flow field based on the intercepting satellite position. The combined potential field of the initial flow field and the disturbed flow field is the planned velocity direction of the satellite. As a decision-making layer, the proximal policy optimization (PPO) dynamically adjusts the perturbed flow field in the IFDS to increase the avoidance success rate in different scenarios. The experimental results show that, compared with the particle swarm optimization with rolling horizon control algorithm, the algorithm proposed in this paper has a shorter decision time and a higher avoidance success rate. At the same time, Monte Carlo simulation shows that the evasion success rate of the proposed algorithm reaches 98%.

Multi-robot systems (MRSs) is a growing field of research that focuses on the collaboration of multiple robots to achieve a common global objective. Managing these systems poses several challenges, including coordination, task allocation, and communication. Among these challenges, a major area of focus is devising an effective communication scheme that ensures robots’ cooperation and adapts to varying conditions during task execution. In this paper, we develop a novel communication management framework tailored for MRSs, specifically addressing dynamic bandwidth distribution in networked teleoperated robotic systems. The algorithm is combined with semi-autonomous formation control based on the Artificial Potential Fields (APF) algorithm, which allows each individual robot to avoid local obstacles autonomously and tries to maintain a desired formation with its neighbors, while the operator is in charge of high-level control only. Common Dynamic Bandwidth Allocation (DBA) algorithms allocate bandwidth to different units based on network conditions and requirements. On the other hand, our proposed DBA scheme dynamically distributes the available bandwidth on communication streams based on factors related to task execution and system performance. In specific, bandwidth is allocated in a way that adapts to changes occurring in the system’s environment and its internal state, including the effect of the autonomous action taken by the path planner on the MRS and the performance of the controller of each individual robot. By addressing the limitations of existing approaches through shaping the communication behavior of the MRS based on performance measures, our proposed algorithm offers a promising solution for improving the performance and efficiency of MRSs. The proposed scheme is tested through simulations on a group of six unmanned aerial vehicles (UAVs) in the Robot Operating System (ROS)-Gazebo simulation environment. The obtained results show the scheme’s capability for enhancing the robotic system’s performance while significantly reducing bandwidth consumption. Experimental testing on two mobile robots further demonstrates the effectiveness of the proposed scheme.

ARIAC is a robotic simulation competition promoted by NIST annually since 2017, aiming to present competitors’ with contemporary industry problems to be solved using agile robotics. For the 2023 competition, ARIAC competitors must perform assembly and kitting tasks by controlling four autonomous ground vehicles (AGVs), one floor-based robot, and one ceiling-based (Gantry) robot in an attempt to overcome a range of agility challenges in the supplied simulated environment, itself based on the Robot Operating System (ROS 2) and Gazebo. The 2023 competition also included a “human” agility challenge, comprising a (simulated) human operator working among robots on the factory floor. This development was motivated by the fact that, while robots and automation play an increasingly significant role in modern manufacturing, there still remains a close relationship between machines and humans. They should complement each other’s strengths and cover each other’s limitations while also observing any required safety rules. For example, the ISO standard “Robots and Robotic Devices – Collaborative robots” (ISO 15066:2016) prescribes the distances required between humans and robots. Within the ARIAC simulation environment, each human operator is controlled using autonomous Belief-Desire-Intention (BDI) agents. At the same time, competitors can monitor the position of each human operator at any time by subscribing to the relevant ROS topic. In this article, we analyse the effects of this (simulated) human presence in the 2023 ARIAC competition and perform a detailed analysis of how the three different human personalities that were implemented affect the assembly tasks undertaken at the four different locations of the assembly stations. Given how the system is currently implemented, it appears that the influence of each encoded personality on the competitors is not as predictable as anticipated. We expand on why this may be a problem when addressing real collaborative spaces involving humans and industrial robots and the improvements that can be undertaken to mitigate the ensuing problems.

The emergence of a new user of the airspace of the Republic of Poland, Unmanned Aerial Vehicles - drones, as part of regular flights, implies the need not only to provide an appropriate legal framework, but also to rebuild the entire air traffic management ecosystem for the integration of manned and unmanned aviation. The first attempts to include UAS in the integrated air traffic system in the Warsaw FIR were based on the electronic/digital planning, coordination and management of UAS flights called Pansa UTM (Polish Air Navigation Services Agency Unmanned Traffic Management). This system is an innovative solution used by the Polish Air Navigation Services Agency (PANSA) and potential drone users. The operation and development of the Pansa UTM system generates the need for air traffic management entities to take specific actions by planning and establishing regulations, rules, safety criteria and flight conditions for the new airspace user. This requires involvement in projects, pilotages and technology demonstrators aimed at creating adequate tools for the safe implementation of UAS flights in Poland. These activities, both already implemented and at the planning stage, aim to achieve a state in which it will be possible to talk about integrated and safe controlled air traffic over Poland. Due to the dynamic nature of the subject matter, this study is only an attempt to present a number of projects undertaken by a wide range of aviation-related entities that may allow achieving an acceptable level of air traffic management over the territory of the Republic of Poland.

Picking up non-trivial objects from a bin with a robotic arm is a common task of modern industrial processes. Here, an efficient data-driven method of grasping point detection, based on an attention squeeze parallel U-shaped neural network (ASP U-Net) for the bin picking task, is proposed. The method directly provides all necessary information about the feasible grasping points of objects, which are randomly or regularly arranged in a bin with side walls. Moreover, the method is able to evaluate and select the optimal grasping point among the feasible ones for two types of end effectors, i.e., a vacuum cup and a parallel gripper. The key element of the utilized ASP U-Net neural network is the transformation of a single RGB-Depth image of the bin containing nontrivial objects into a schematic grey-scale frame, where the positions and poses of the grasping points are coded into gradient geometric shapes. The experiments carried out in this study include a comprehensive set of scenes with randomly scattered, ordered, and semi-ordered objects arranged in impeccable or deformed bins. The results indicate outstanding accuracy with more than acceptable computational requirements. Additionally, the scaling possibilities of the method can offer extremely lightweight implementations, applicable, for example, to battery-powered edge-computing devices with low RAM capacity.

Deep Reinforcement Learning applications are growing due to their capability of teaching the agent any task autonomously and generalizing the learning. However, this comes at the cost of a large number of samples and interactions with the environment. Moreover, the robustness of learned policies is usually achieved by a tedious tuning of hyper-parameters and reward functions. In order to address this issue, this paper proposes an evolutionary RL algorithm for the adaptive optimization of hyper-parameters. The policy is trained using an on-policy algorithm, Proximal Policy Optimization (PPO), coupled with an evolutionary algorithm. The achieved results demonstrate an improvement in the sample efficiency of the RL training on a robotic grasping task. In particular, the learning is improved with respect to the baseline case of a non-evolutionary agent. The evolutionary agent needs (60)% fewer samples to completely learn the grasping task, enabled by the adaptive transfer of knowledge between the agents through the evolutionary algorithm. The proposed approach also demonstrates the possibility of updating reward parameters during training, potentially providing a general approach to creating reward functions.

Future autonomous Unmanned Aerial Vehicles (UAV) missions will take place in highly cluttered urban environments. As a result, the UAV must be able to autonomously evaluate risks and react to unforeseen hazards. The current regulatory framework for missions implements SORA guidelines for hazard detection, but its application to air-to-air collision is limited. This research defined a rigorous verification and validation framework (V&V) for digital twins for use in future autonomous UAV missions. The researchers designed a sentry mission for a UAV to evaluate its capacity to detect small uncooperative flying objects. A digital twin of the DJI M300 vision system was built using a game engine and a V&V framework was developed to assure the quality of results in both virtual and real-world scenarios. The results showed the capability of the digital twin to identify vulnerabilities and worst-case scenarios in UAV mission operations, and how it can assist remote pilots in identifying air-to-air collision hazards. Furthermore, the probability of air-to-air collision was calculated for three sentry patterns, and the results were validated in the field. This research demonstrated the capability to identify vulnerabilities and worst-case scenarios in UAV mission operations. We present how the digital twin of an operational theatre can be exploited to assist remote pilots with the identification of air-to-air collision hazards of small uncooperative objects. Furthermore, we discuss how these results can be used to enhance current SORA-based risk assessment practices.

Human-Robot Interaction is an increasingly important topic in both research and industry fields. Since human safety must be always guaranteed and accidental contact with the operator avoided, it is necessary to investigate real-time obstacle avoidance strategies. The transfer from simulation environments, where algorithms are tested, to the real world is challenging from different points of view, e.g., the continuous tracking of the obstacle and the configuration of different manipulators. In this paper, the authors describe the implementation of a collision avoidance strategy based on the potential field method for off-line trajectory planning and on-line motion control, paired with the Motion Capture system Optitrack PrimeX 22 for obstacle tracking. Several experiments show the performance of the proposed strategy in the case of a fixed and dynamic obstacle, disturbing the robot’s trajectory from multiple directions. Two different avoidance modalities are adapted and tested for both standard and redundant robot manipulators. The results show the possibility of safely implementing the proposed avoidance strategy on real systems.

The RRT*-Connect algorithm enhances efficiency through dual tree bias growth, yet this bias can be inherently blind, potentially affecting the algorithm’s heuristic performance. In contrast, the Informed RRT* algorithm narrows the planning problem’s scope by leveraging an informed region, thereby improving convergence efficiency towards optimal solutions. However, this approach relies on the prior establishment of feasible paths. Combining these two algorithms can address the challenges posed by Informed RRT while also accelerating convergence towards optimality, albeit without resolving the issue of blind bias in dual trees.In this paper, we proposed a novel algorithm: Dynamic Informed Bias RRT*-Connect. This algorithm, grounded in potential and explicit informed bias sampling, introduces a dynamical bias points set that guides dual tree growth with precision objectives. Additionally, we enhance the evaluation framework for algorithmic heuristics by introducing two innovative metrics that effectively capture the algorithm’s characteristics. The improvements observed in traditional indicators demonstrate that the proposed algorithm exhibits greater heuristic compared to RRT*-Connect and Informed RRT*-Connect. These findings also suggest the viability of the new metrics introduced in our evaluation framework.

The spherical robot XK-III, designed with redundant degrees of freedom, addresses the limitations of existing pendulum spherical robot structures by enhancing mobility and environmental adaptability. A nonlinear dynamic model is developed for XK-III’s new drive structure, along with a nonlinear disturbance observer (NDOB) to mitigate perturbations. Additionally, a Fuzzy PID controller (FPID) is implemented to further enhance XK-III’s environmental adaptability. Experimental results confirm the effectiveness of the new design, showing that XK-III equipped with FPID and NDOB outperforms traditional control systems in terms of anti-disturbance capabilities. This research provides valuable insights for the use of spherical robots in complex environments.