Task planning for human-robot collaboration in structure assembly

Studies in the literature on flexible job-shop scheduling problems (FJSP) generally assume that one worker is assigned to each machine and that processing times are constant. However, in some industries, multiple workers with cooperation can process complex operations faster than one worker. If the possibility of completing jobs in a shorter time with worker cooperation is not taken into account, the opportunity to create more effective schedules may not be taken advantage of. Therefore, it is essential to consider the flexibility of collaboration between employees. However, to increase labor efficiency in businesses, jobs are also expected to be done with the minimum number of workers possible. This study considers the FJSP with both machine and number of workers dependent processing times. The objectives are minimizing the total tardiness and the total number of workers. A bi-objective mathematical model and an NSGA-II algorithm for large-sized problems have been proposed. The performance of the proposed solution approaches is demonstrated by using randomly generated test problems. For each problem, the most successful Pareto solution among the obtained solutions by the mathematical model and the NSGA-II algorithm was determined using the TOPSIS method. Furthermore, the effect of the total number of workers on the total tardiness is examined. The performance of proposed solution approaches, and when the worker number increases, the total tardiness of jobs can be reduced by an average of 75.88%, have been shown through comprehensive experimental studies.

Continuous timely repair and replacement of infrastructures, equipment and utilities play an important role in maintaining the smooth-running of a city or local community. Thereby, to help individuals and businesses go about their daily activities with ease, it is vital to develop a proper method for automatically identifying and assigning capable workers for tasks. This paper defines the community management service task allocation problem as CMS-TAP and hence an end-to-end “recommendation + allocation” network, i.e. a task recommender and allocation optimization network (denoted as TROpt-NET), is then developed for handling such problem. TROpt-NET consists of two layers, namely one for predicting worker ability and the other for allocating tasks which are TR Layer and TA Layer, corresponding to “recommendation” and “allocation” of tasks. Different from operations research approaches where workers are assigned to jobs based on their pre-labelled skills and fixed locations, we propose a task recommender and allocation optimization network. The TR layer is a task recommender system designed to learn implicit worker abilities for different tasks using Neural Collaborative Filtering (NCF) by mining a historical dataset of worker task completion. Whereas in the TA layer a differential optimization approach for allocation is used because of its differentiable property and ability to allow for backpropagation to the prediction layer. In this study, we first formulate the CMS-TAP problem as a recommendation +optimization problem and then propose and end-to-end network architecture that tackles the problem in a real-world setting. TROpt-NET curbs uncertainty and assumptions in optimization by learning to more accurately approximate worker ability across different tasks. Additionally, the network can learn implicit worker abilities enabling optimal utilization of workers across a wide range of tasks, which is often ignored in task allocation problems. We find that normalizing worker ability across all tasks improves the implicit learning capability of the network and that good approximations don’t always lead to optimal allocation but learning allocations by backpropagating through recommendations improves the allocation objective. Offline experiments on a real-world large-scale dataset demonstrate the effectiveness of our proposed TROpt-NET.

Human-robot collaboration (HRC), as a part of Industry 4.0 strategy, requires a completely new type of robots able to co-work with humans, called collaborative robots or cobots. This kind of collaboration is especially needed in assembly systems, which are known for having a low level of automation. For some assembly tasks human is still an irreplaceable factor. On the other hand, some assembly tasks are monotonous and tiring for humans. Therefore, the different approaches to cope with the challenge of identification and selection of proper task allocation between human worker and cobots are reported by many researchers. It is not an easy task since multiple and often conflicting criteria need to be taken into account. Some kind of artificial decision support is needed, to successfully solve this problem. In this research, task allocation procedure is presented for identification of different improvement options that utilize cobots into the assembly line for different tasks to be performed. The decision support system based on the HUMANT algorithm has been used for selection of the option which represents the best compromise solution. The procedure is experimentally tested on the assembly line with car gearboxes as a real product.

In order to minimize the processing time for multiple robots to finish lots of tasks, an optimized multiple robots task allocation and path planning approach is proposed. In this approach, the optimized task sequence is first schemed on the assumption that all the tasks are processed by one robot. While scheming the task sequence. A star algorithm is employed to plan paths and calculate the path cost. Then, according to the number of the robots, the task sequence has been randomly divided into several task segments that will be assigned to robots. At last, the task number of each task segments are averaged according to the time each robot used, so that the optimized task allocation scheme is got. In this approach, not only multiple robot task allocation, but also the optimized path planning is realized to ensure that the robots can finish all the tasks in the shortest time. The experiments and simulation are provided to validate the effectiveness of the approach. The results show that the proposed approach can scheme optimized multiple robots task allocation and path planning scheme.

In the logistics robot picking system, how to assign tasks to each robot reasonably determines the running time and cost of the whole system. A robust optimization model for task allocation based on logistics robot picking system is presented in this paper. In the model, the task completion time is taken as the objective function, and the actual walking distance of the robot is used as the uncertainty factor. With the robust optimization theory, the robust peer-to-peer transformation was performed on the task allocation model with uncertain parameters. Then, the task allocation model is solved by the Hungary algorithm. Finally, the effectiveness of the model is verified by an example simulation. The optimized task completion time is not sensitive to the uncertainty distance and is relatively stable.

In many current assembly scenarios, efficient collaboration between humans and robots can improve collaborative efficiency and quality. However, the efficient arrangement of human–robot collaborative (HRC) tasks constitutes a significant challenge. In a collaborative workspace where humans and robots collaborate on assembling a shared product, the determination of task allocation between them is of crucial importance. To address this issue, offline feasible HRC paths are established based on assembly task constraint information. Subsequently, the HRC process is simulated within a virtual environment leveraging these feasible paths. Human assembly intentions are explicitly expressed through human assembly trajectories, and implicitly expressed through simulation results such as assembly time and human–robot resource allocation. Furthermore, a two-stage MLP-LSTM network is employed to train and optimize the assembly simulation database. In the first stage, a sequence generation model is trained using high-quality HRC processes. Then, the network learns human evaluation patterns to score the generated sequences. Ultimately, task allocation for HRC is performed based on the high-scoring generated sequences. The effectiveness of the proposed method is demonstrated through assembly scenarios of two products. Compared with traditional optimization methods like DFS and Greedy, the human collaboration ratio has been optimized by 10%, while the collaborative quality evaluation has been improved by 3%.

Migrant workers are a unique group under China's urban-rural dual system, and improving their well-being is an intrinsic requirement for achieving common prosperity. In recent years, human-robot collaboration has come forth as a novel work paradigm. Comprehending the influence of human-robot collaboration on the well-being of migrant workers is a vital yet insufficiently investigated matter. With the conservation of resources theory as an analytical framework, this study empirically analyzes data from a two-stage survey of 382 migrant workers in Chinese manufacturing enterprises. The findings suggest that perceived human-robot collaboration can positively influence migrant workers' well-being by facilitating perceived decent work, while also negatively affecting their well-being by increasing job replacement anxiety. Perceived organizational support plays a positive moderating role in two distinct aspects: on the one hand, the relationship between perceived human-robot collaboration and perceived decent work; on the other hand, the mediating effect through which perceived decent work connects perceived human-robot collaboration with the well-being of migrant workers. Conversely, perceived organizational support exerts a negative moderating effect on the association between perceived human-robot collaboration and job replacement anxiety, while job replacement anxiety functions as a mediator in the relationship linking perceived human-robot collaboration to migrant workers' well-being. This study helps understand how human-robot collaboration in manufacturing enterprises affects the well-being of migrant workers.

The challenge concerning the optimal allocation of tasks across multiple unmanned aerial vehicles (multi-UAVs) has significantly spurred research interest due to its contribution to the success of various fleet missions. This challenge becomes more complex in time-constrained missions, particularly if they are conducted in hostile environments, such as search and rescue (SAR) missions. In this study, a novel fish-inspired algorithm for multi-UAV missions (FIAM) for task allocation is proposed, which was inspired by the adaptive schooling and foraging behaviors of fish. FIAM shows that UAVs in an SAR mission can be similarly programmed to aggregate in groups to swiftly survey disaster areas and rescue-discovered survivors. FIAM’s performance was compared with three long-standing multi-UAV task allocation (MUTA) paradigms, namely, opportunistic task allocation scheme (OTA), auction-based scheme, and ant-colony optimization (ACO). Furthermore, the proposed algorithm was also compared with the recently proposed locust-inspired algorithm for MUTA problem (LIAM). The experimental results demonstrated FIAM’s abilities to maintain a steady running time and a decreasing mean rescue time with a substantially increasing percentage of rescued survivors. For instance, FIAM successfully rescued 100% of the survivors with merely 16 UAVs, for scenarios of no more than eight survivors, whereas LIAM, Auction, ACO and OTA rescued a maximum of 75%, 50%, 35% and 35%, respectively, for the same scenarios. This superiority of FIAM performance was maintained under a different fleet size and number of survivors, demonstrating the approach’s flexibility and scalability.

In this paper, we consider the task allocation problem for autonomous agents where the distances between any two agents should always be less than a predetermined value. To easily tackle the problem, we introduce a novel concept referred to as a virtual task that helps to decouple the task allocation from the path planning in the joint task allocation and path planning problem. An overall task allocation procedure with two steps is proposed. After the virtual tasks are generated with preliminary clustering in the first step, tasks are clustered into groups and allocated to agents in the second step. To reduce the computational complexity of optimal clustering and task allocation, we propose a suboptimal method called the cluster-level traveling salesman problem (TSP). Numerical results show that the cluster-level TSP suffers only little performance degradation compared to the optimal method, even though its computational costs are dramatically reduced.

How to guarantee the data rate and latency requirement for an application with limited energy is an open issue in wireless virtualized sensor networks. In this paper, we integrate the wireless energy transfer technology into the wireless virtualized sensor network and focus on the stochastic performance guarantee. Firstly, a joint task and resource allocation optimization problem are formulated. In order to characterize the stochastic latency of data transmission, effective capacity theory is resorted to study the relationship between network latency violation probability and the transmission capability of each node. The performance under the FDMA mode and that under the TDMA mode are first proved to be identical. We then propose a bisection search approach to ascertain the optimal task allocation with the objective to minimize the application latency violation probability. Furthermore, a one-dimensional searching scheme is proposed to find out the optimal energy harvesting time in each time block. The effectiveness of the proposed scheme is finally validated by extensive numerical simulations. Particularly, the proposed scheme is able to lower the latency violation probability by 11.6 times and 4600 times while comparing with the proportional task allocation scheme and the equal task allocation scheme, respectively.

Task allocation and scheduling models for distributed digital signal processing are presented. The notions of block-type and stream-type tasks in signal processing application are introduced, and models for sequential and parallel I/O are presented. By extending the traditional models, more accurate schedules can be obtained. Those models can be further enhanced by allowing additional restrictions on the number of parallel I/O ports and the amount of parallelism on memory access. The deterministic nature of digital signal processing algorithms allows for more computationally intensive and accurate task allocation techniques to be performed at compile time. By applying a branch and bound algorithm, the task allocation problem can easily be solved for a variety of scheduling models and various system restrictions.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A typical mission consists of several deadline-driven, inter-dependent tasks which need to adhere to specific resource constraints. This makes Optimal Task Allocation (OTA) of a particular mission a daunting task, even for simple and less-challenging environments. The inherent challenges of Mission Critical Environment (MCE) (e.g., Disconnected, intermittent, and limited communication among nodes, high error rate, node mobility, etc.) make OTA even harder to accomplish. To address this problem, earlier, we proposed an automated approach, namely, Centralized Optimal Task Allocation Algorithm (COTAA). While COTAA performs OTA in an efficient manner, it is based on some assumptions (e.g., nodes must follow publish/subscribe-based communication model, there is no inter-dependency among tasks, and static central unit is solely responsible for task allocation) that make COTAA applicable only to specific MCEs (e.g., post-disaster recovery) and do not work well for other MCEs such as group-based UAV (Unmanned Aerial Vehicle) operation, robot-driven mission, etc. In this paper, we propose another novel automated approach, Decentralized Optimal Task Allocation Algorithm (DOTAA), which relaxes the above-mentioned assumptions and performs OTA in an efficient manner. In DOTAA, we have exploited the concept of application-layer hash and bidding approach to perform the OTA for larger classes of MCEs. We have also evaluated our solution using ns-2 simulator and our results show that DOTAA outperforms COTAA in terms of scalability, task allocation time, and bandwidth consumption.

In radar networks, research on the task allocation strategy for multi-target imaging is essential, especially under limited radar resources. From the targets' perspective, the goal is to acquire enough resources to obtain high-quality inverse synthetic aperture radar (ISAR) images. Then there is competition among targets for radar resources. Therefore, the task allocation optimization problem is considered as a cooperative game, specifically a non-transferable coalitional game when the sparse ISAR imaging algorithm is adopted. A stable coalition structure, which allows cooperation among targets to achieve satisfactory image resolution using the least amount of time, represents an optimal task allocation scheme. Then, the corresponding algorithm based on coalition-forming rules is proposed. Through the framework of cooperative game theory, a stable network structure (i.e., the task allocation strategy) is formed while minimizing the radar time resource cost. Finally, experiments confirm that the proposed algorithm can achieve an optimal task allocation strategy with a lower computation time cost.

Multirobot systems (MRS) hold the promise of improved performance and increased fault tolerance for large-scale problems. One of the most important aspects in the design of MRS is the allocation of tasks among the robots in a productive and efficient manner. Optimal solutions to multirobot task allocation (MRTA) can be found through an exhaustive search. Since there are ways in which m tasks can be assigned to n robots, an exhaustive search is often not possible. Task allocation methodologies must ensure that not only the global mission is achieved, but also the tasks are well distributed among the robots. This paper presents task allocation methodologies for MRS by considering their capability in terms of time and space. A two-phase solution methodology is used to solve the MRTA problem wherein the task capacity of the robots is determined during the first phase and the task allocation optimization is done during the second phase using linear programming (LP).

Multi-UAV reconnaissance task allocation for heterogeneous targets using an opposition-based genetic algorithm with double-chromosome encoding