3D Reactive Control and Frontier-Based Exploration for Unstructured Environments

Ahmad, Shakeeb; Mills, Andrew B.; R., Rush, Eugene; Frew, Eric W.; Humbert, J. Sean

doi:10.1109/iros51168.2021.9636575

Cited by 10 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In case of a UAV navigating in such an environment, even a single close contact of the robot with an obstacle, for instance an electric cable, can prove detrimental to the exploration mission safety. Our two-tier planning setup demonstrated safe and reliable exploration through the environment (Ahmad et al, 2021a). One of the runs is highlighted in Figure 17 in which the UAV navigated around 300 m with a full battery pack which amounted for 11 minutes of flight time.…”

Section: Aerial Vehicle Vision-based Local Controlmentioning

confidence: 99%

Multi-Agent Autonomy: Advancements and Challenges in Subterranean Exploration

Ohradzansky

Riley

et al. 2022

Self Cite

View full text Add to dashboard Cite

Artificial intelligence has undergone immense growth and maturation in recent years, though autonomous systems have traditionally struggled when fielded in diverse and previously unknown environments. DARPA is seeking to change that with the Subterranean Challenge, by providing roboticists the opportunity to support civilian and military first responders in complex and high-risk underground scenarios. The subterranean domain presents a handful of challenges, such as limited communication, diverse topology and terrain, and degraded sensing. Team MARBLE proposes a solution for autonomous exploration of unknown subterranean environments in which coordinated agents search for artifacts of interest. The team presents two navigation algorithms in the form of a metric-topological graph-based planner and a continuous frontier-based planner. To facilitate multi-agent coordination, agents share and merge new map information and candidate goal points. Agents deploy communication beacons at different points in the environment, extending the range at which maps and other information can be shared. Onboard autonomy reduces the load on human supervisors, allowing agents to detect and localize artifacts and explore autonomously outside established communication networks. Given the scale, complexity, and tempo of this challenge, a range of lessons was learned, most importantly, that frequent and comprehensive field testing in representative environments is key to rapidly refining system performance.

show abstract

Section: Aerial Vehicle Vision-based Local Controlmentioning

confidence: 99%

Multi-Agent Autonomy: Advancements and Challenges in Subterranean Exploration

Ohradzansky

Riley

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…The use of spatial generation of pyramids of the free spaces, allows for labeling obstacle free trajectories that lie inside the pyramids, while achieving fast generation of large number of candidate trajectories and performs collision checks. In [10] the authors present a reactive navigation system for MAV exploration. The developed algorithm is based on a two layered planning architecture that leverages the global environment map for frontier generation and local instantaneous sensor data for obstacle avoidance based on artificial potential fields.…”

Section: A State Of the Artmentioning

confidence: 99%

Fast Planner for MAV Navigation in Unknown Environments Based on Adaptive Search of Safe Look-Ahead Poses

Patel

Lindqvist

Kanellakis

et al. 2022

2022 30th Mediterranean Conference on Control and Automation (MED)

View full text Add to dashboard Cite

Autonomous navigation capability is a crucial part for deploying robots in an unknown environment. In this article a reactive local planner for autonomous and safe navigation in subterranean environment is presented. The proposed planning framework navigates the MAV forward in a tunnel such that the MAV gains more information about the environment while avoiding obstacles. The proposed planning architecture works solely based on the information of local surrounding of the MAV thus, making navigation simple yet fast. One of the biggest novelties of the article comes from solving the combined problem of autonomous navigation and obstacle avoidance. The proposed algorithm for selecting the next way point of interest also accounts in the safety margin for traversing to such way point. The approach presented in this article is also different from classical map based global planning algorithms because it favours the next way point away from obstacles in way point selection process and thus providing a safe path for incremental forward navigation. The approach is validated by simulating a MAV equipped with the proposed reactive local planner in order for the MAV to navigate in a subterranean cave environment.

show abstract

“…The use of spatial generation of pyramids of the free spaces, allows for labeling obstacle free trajectories that lie inside the pyramids, while achieving fast generation of large number of candidate trajectories and performs collision checks. In [2] the authors present a reactive navigation system for MAV exploration. The developed algorithm is based on a two layered planning architecture that leverages the global environment map for frontier generation and local instantaneous sensor data for obstacle avoidance based on artificial potential fields.…”

Section: Related Workmentioning

confidence: 99%

REF: A Rapid Exploration Framework for Deploying Autonomous MAVs in Unknown Environments

Patel¹,

Lindqvist²,

Kanellakis³

et al. 2022

Preprint

View full text Add to dashboard Cite

Exploration and mapping of unknown environments is a fundamental task in applications for autonomous robots. In this article, we present a complete framework for deploying MAVs in autonomous exploration missions in unknown subterranean areas. The main motive of exploration algorithms is to depict the next best frontier for the robot such that new ground can be covered in a fast, safe yet efficient manner. The proposed framework uses a novel frontier selection method that also contributes to the safe navigation of autonomous robots in obstructed areas such as subterranean caves, mines, and urban areas. The framework presented in this work bifurcates the exploration problem in local and global exploration. The proposed exploration framework is also adaptable according to computational resources available onboard the robot which means the trade-off between the speed of exploration and the quality of the map can be made. Such capability allows the proposed framework to be deployed in a subterranean exploration, mapping as well as in fast search and rescue scenarios. The overall system is considered a low-complexity and baseline solution for navigation and object localization in tunnel-like environments. The performance of the proposed framework is evaluated in detailed simulation studies with comparisons made against a high-level explorationplanning framework developed for the DARPA Sub-T challenge as it will be presented in this article.

show abstract

3D Reactive Control and Frontier-Based Exploration for Unstructured Environments

Cited by 10 publications

References 25 publications

Multi-Agent Autonomy: Advancements and Challenges in Subterranean Exploration

Multi-Agent Autonomy: Advancements and Challenges in Subterranean Exploration

Fast Planner for MAV Navigation in Unknown Environments Based on Adaptive Search of Safe Look-Ahead Poses

REF: A Rapid Exploration Framework for Deploying Autonomous MAVs in Unknown Environments

Contact Info

Product

Resources

About