Imitation learning for task allocation

Duvallet, Felix; Stentz, Anthony

doi:10.1109/iros.2010.5650006

Cited by 17 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent structured prediction approach [37] for task allocation uses a combination of reinforcement learning and quadratic integer programming for learning directly from data to optimize assignments. The approaches in [17], [37], however, assume the existence of a single strategy for task allocation. A distributed approach for multi-agent task allocation [38] learns to select the most appropriate of two pre-specified strategies, namely Earliest Deadline First (EDF) or Nearest Task First (NTF).…”

Section: Learning For Team-level Coordinationmentioning

confidence: 99%

“…The learning methods discussed so far consider only homogeneous robots [37], do not show generalization to teams unseen during training [17], [18], depend on the ability to interact with the environment to learn policies [39] or adhere to a limited number of pre-specified strategies [38]. In contrast, our framework is capable of learning generalizable and heterogeneous strategies for task allocation in heterogeneous multi-agent systems from expert demonstrations, and do not rely on environmental interactions.…”

Section: Learning For Team-level Coordinationmentioning

confidence: 99%

See 1 more Smart Citation

Resource-Aware Adaptation of Heterogeneous Strategies for Coalition Formation

Anusha¹,

Ravichandar²

2021

Preprint

View full text Add to dashboard Cite

Existing approaches to coalition formation often assume that requirements associated with tasks are precisely specified by the human operator. However, prior work has demonstrated that humans, while extremely adept at solving complex problems, struggle to explicitly state their solution strategy. In this work, we propose a framework to learn implicit task requirements directly from expert demonstrations of coalition formation. We also account for the fact that demonstrators may utilize different, equally-valid solutions to the same task. Essentially, we contribute a framework to model and infer such heterogeneous strategies to coalition formation. Next, we develop a resource-aware approach to generalize the inferred strategies to new teams without requiring additional training. To this end, we formulate and solve a constrained optimization problem that simultaneously selects the most appropriate strategy for a given target team, and optimizes the constituents of its coalitions accordingly. We evaluate our approach against several baselines, including some that resemble existing approaches, using detailed numerical simulations, StarCraft II battles, and a multi-robot emergencyresponse scenario. Our results indicate that our framework consistently outperforms all baselines in terms of requirement satisfaction, resource utilization, and task success rates.

show abstract

Section: Learning For Team-level Coordinationmentioning

confidence: 99%

Section: Learning For Team-level Coordinationmentioning

confidence: 99%

Resource-Aware Adaptation of Heterogeneous Strategies for Coalition Formation

Anusha¹,

Ravichandar²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Reinforcement learning is an early used architecture in this field [20]. Meanwhile, imitation learning is also used [21], which is characterized by expert's knowledge. Nunes and Gini study an auction algorithm to allocate temporal-constraint tasks, and model the temporal constraints on the tasks as a simple temporal problem [22].…”

Section: Related Work a Multi-robot Task Allocationmentioning

confidence: 99%

Multi-Robot Cooperative Task Allocation With Definite Path-Conflict-Free Handling

et al. 2019

View full text Add to dashboard Cite

Modeling and solving multi-robot task allocation with definite path-conflict-free handling is an important research, especially in real working environments. Some of the research lines are unable to obtain definite path-conflict-free solutions for multi-robot task allocations, such as using the penalty-term method in the fitness function to restrict the survival probabilities of the solutions with path conflicts. In some cases, these solutions are only able to satisfy the objective of minimizing task time. We formulate this problem based on grid maps, while focusing on the frequently used cooperative task allocation. In our model, two subtasks of each cooperative task must be executed by two robots, simultaneously. We propose vitality-driven genetic task allocation algorithm (VGTA), which is able to simultaneously minimize task time and realize definite conflict-free path planning. VGTA consists of local operators, such as random mutations, greedy crossovers, and vitality selection. Meanwhile, VGTA includes schedule conflict and path conflict handling strategies. In path conflict handling strategy, we not only consider the common path conflicts in a grid cell, but also focus on the path conflicts between robots when exchanging positions in the adjacent grid cells. Besides, we construct our benchmarks based on real working environments, such as factory, powerhouse, and airport environments. Experimental results indicate that VGTA's search capability and computation cost are satisfactory. Meanwhile, its solutions are able to be really executed.INDEX TERMS Multi-robot task allocation, cooperative task, schedule conflict, path conflict, unmanned multi-robot swarm.

show abstract

“…In the past decade, multirobot task allocation has been a popular research topic in robotics. Furthermore, a few learning‐based multirobot task allocation approaches have been developed for such applications as fire‐fighting disaster response (Duvallet and Stentz, ; Jones et al., ), patrolling (Tangamchit et al., ), and multirobot auctioning (Pippin and Christensen, ).…”

Section: Related Workmentioning

confidence: 99%

“…In Duvallet and Stentz (), imitation learning was implemented to incorporate human expert knowledge into the decision‐making process of a market‐based multirobot task allocation method for fire‐fighting disaster response. Namely, the human expert's solutions to a multirobot task allocation problem were represented by a set of demonstrated allocations that were then utilized to train a pricing policy, e.g., assign different values to different buildings.…”

Section: Related Workmentioning

confidence: 99%

Multirobot Cooperative Learning for Semiautonomous Control in Urban Search and Rescue Applications

Liu

Nejat

2015

Journal of Field Robotics

View full text Add to dashboard Cite

The use of cooperative multirobot teams in urban search and rescue (USAR) environments is a challenging yet promising research area. For multirobot teams working in USAR missions, the objective is to have the rescue robots work effectively together to coordinate task allocation and task execution between different team members in order to minimize the overall exploration time needed to search disaster scenes and to find as many victims as possible. This paper presents the development of a multirobot cooperative learning approach for a hierarchical reinforcement learning (HRL) based semiautonomous control architecture in order to enable a robot team to learn cooperatively to explore and identify victims in cluttered USAR scenes. The proposed cooperative learning approach allows effective task allocation among the multirobot team and efficient execution of the allocated tasks in order to improve the overall team performance. Human intervention is requested by the robots when it is determined that they cannot effectively execute an allocated task autonomously. Thus, the robot team is able to make cooperative decisions regarding task allocation between different team members (robots and human operators) and to share experiences on execution of the allocated tasks. Extensive results verify the effectiveness of the proposed HRL‐based methodology for multi‐robot cooperative exploration and victim identification in USAR‐like scenes.

show abstract

Imitation learning for task allocation

Cited by 17 publications

References 14 publications

Resource-Aware Adaptation of Heterogeneous Strategies for Coalition Formation

Resource-Aware Adaptation of Heterogeneous Strategies for Coalition Formation

Multi-Robot Cooperative Task Allocation With Definite Path-Conflict-Free Handling

Multirobot Cooperative Learning for Semiautonomous Control in Urban Search and Rescue Applications

Contact Info

Product

Resources

About