Pooyan Fazli

Complete and robust cooperative robot area coverage with limited range

We address the problem of multi-robot area coverage and present a new approach in the case where the map of the area and its static obstacles are known and the robots have a limited visibility range. The proposed method starts by locating a set of static guards on the map of the target area and then builds a graph called Reduced-CDT, a new environment representation method based on the Constrained Delaunay Triangulation (CDT). Multi-Prims is used to decompose the graph into a forest of partial spanning trees (PSTs). Each PST is then modified through a mechanism called Constrained Spanning Tour (CST) to build a cycle which is then assigned to an individual robot. Subsequently, robots start navigating the cycles and consequently cover the whole area. We show that the proposed approach is complete and robust with respect to robot failure.

The 2014 International Planning Competition: Progress and Trends

We review the 2014 International Planning Competition (IPC-2014), the eighth in a series of competitions starting in 1998. IPC-2014 was held in three separate parts to assess state-of-the-art in three prominent areas of planning research: the deterministic (classical) part (IPCD), the learning part (IPCL), and the probabilistic part (IPPC). Each part evaluated planning systems in ways that pushed the edge of existing planner performance by introducing new challenges, novel tasks, or both. The competition surpassed again the number of competitors than its predecessor, highlighting the competition’s central role in shaping the landscape of ongoing developments in evaluating planning systems.

Multi-robot repeated area coverage

We address the problem of repeated coverage of a target area, of any polygonal shape, by a team of robots having a limited visual range. Three distributed Cluster-based algorithms, and a method called Cyclic Coverage are introduced for the problem. The goal is to evaluate the performance of the repeated coverage algorithms under the effects of the variables: Environment Representation, and the Robots’ Visual Range. A comprehensive set of performance metrics are considered, including the distance the robots travel, the frequency of visiting points in the target area, and the degree of balance in workload distribution among the robots. The Cyclic Coverage approach, used as a benchmark to compare the algorithms, produces optimal or near-optimal solutions for the single robot case under some criteria. The results can be used as a framework for choosing an appropriate combination of repeated coverage algorithm, environment representation, and the robots’ visual range based on the particular scenario and the metric to be optimized.

Curious George: An Integrated Visual Search Platform

This paper describes an integrated robot system, known as Curious George, that has demonstrated state-of-the-art capabilities to recognize objects in the real world. We describe the capabilities of this system, including: the ability to access web-based training data automatically and in near real-time, the ability to model the visual appearance and 3D shape of a wide variety of object categories, navigation abilities such as exploration, mapping and path following, the ability to decompose the environment based on 3D structure, allowing for attention to be focused on regions of interest, the ability to capture high-quality images of objects in the environment, and finally, the ability to correctly label those objects with high accuracy. The competence of the combined system has been validated by entry into an international competition where Curious George has been among the top performing systems each year. We discuss the implications of such successful object recognition for society, and provide several avenues for potential improvement.

Multi-robot repeated boundary coverage under uncertainty

We address the problem of repeated coverage by a team of robots of the boundaries of a target area and the structures inside it. Events may occur on any parts of the boundaries and may have different importance weights. In addition, the boundaries of the area and the structures are heterogeneous, so that events may appear with varying probabilities on different parts of the boundary, and this probability may change over time. The goal is to maximize the reward by detecting the maximum number of events, weighted by their importance, in minimum time. The reward a robot receives for detecting an event depends on how early the event is detected. To this end, each robot autonomously and continuously learns the pattern of event occurrence on the boundaries over time, capturing the uncertainties in the target area. Based on the policy being learned to maximize the reward, each robot then plans in a decentralized manner to select the best path at that time in the target area to visit the most promising parts of the boundary. The performance of the learning algorithm is compared with a heuristic algorithm for the Travelling Salesman Problem, on the basis of the total reward collected by the team during a finite repeated boundary coverage mission.

On multi-robot area coverage

Area coverage is one of the emerging problems in multi-robot coordination. In this task a team of robots is cooperatively trying to observe or sweep an entire area, possibly containing obstacles, with their sensors or actuators. The goal is to build an efficient path for each robot which jointly ensure that every single point in the environment can be seen or swept by at least one of the robots while performing the task.

On multi-robot area and boundary coverage

Distributed coverage aims to deploy a team of robots to move around a target area to perform sensing, monitoring, data collection, search, or distributed servicing tasks. This thesis investigates three variations of the coverage problem. First, we address the multi-robot single coverage of a target area. The aim is to guarantee that every accessible point in the area is visited in a finite time. The proposed algorithm supports heterogeneous robots having various maximal speeds, and is robust to robot failure. It also balances the workload distribution among the robots based on their maximal speeds. The obtained results on the coverage time are scalable to workspaces of different sizes, and robots of varied visual ranges. Second, we tackle the multi-robot repeated coverage of a target area. The objective is to visit all the accessible points of the area repeatedly over time, while optimizing some performance criteria. We introduce four repeated coverage algorithms, and evaluate them under a comprehensive set of metrics including the sum of the paths/tours generated for the robots, the frequency of visiting the points in the target area, and the degree of balance in workload distribution among the robots. We also investigate the effects of environment representation, and the robots’ visual range on the performance of the proposed algorithms. The results can be used as a framework for choosing an appropriate combination of repeated coverage algorithm, environment representation, and the robots’ visual range based on the particular workspace and the metric to be optimized. Third, we focus on the multi-robot repeated coverage of the boundaries of a target area and the structures inside it. Events may occur at any position on the boundaries, and the robots are not a priori aware of the event distribution. The goal is to maximize the total detection reward of the events. The reward a robot

Safe navigation in dynamic, unknown, continuous, and cluttered environments

We introduce ProbLP, a probabilistic local planner, for safe navigation of an autonomous robot in dynamic, unknown, continuous, and cluttered environments. We combine the proposed reactive planner with an existing global planner and evaluate the hybrid in challenging simulated environments. The experiments show that our method achieves a 77% reduction in collisions over the straight-line local planner we use as a benchmark.

Help Me! Sharing of Instructions Between Remote and Heterogeneous Robots

Service robots frequently face similar tasks. However, they are still not able to share their knowledge efficiently on how to accomplish those tasks. We introduce a new framework, which allows remote and heterogeneous robots to share instructions on the tasks assigned to them. This framework is used to initiate tasks for the robots, to receive or provide instructions on how to accomplish the tasks, and to ground the instructions in the robots’ capabilities. We demonstrate the feasibility of the framework with experiments between two geographically distributed robots and analyze the performance of the proposed framework quantitatively.