Control-driven mapping and planning

Wooden, David; Powers, Matthew D.; MacKenzie, Doug; Balch, Tucker; Egerstedt, Magnus

doi:10.1109/iros.2007.4399197

Cited by 6 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Careful tuning of the weights of each controller's votes yields a behavior that drives toward a goal location while avoiding obstacles and is fairly robust in avoiding collisions. This hand-crafted behavior is described in detail in Wooden, Powers, MacKenzie, Balch, and Egerstedt (2007).…”

Section: Experiments and Resultsmentioning

confidence: 99%

Learning outdoor mobile robot behaviors by example

Pippin¹,

Balch²

2009

Journal of Field Robotics

View full text Add to dashboard Cite

We present an implementation and analysis of a real-time, online, supervised learning system for nonparametrically learning behaviors from a human trainer on a mobile robot in outdoor environments. This approach enables a human operator to train and tune robot behaviors simply by driving the robot with a remote control. Hand-designed behaviors for outdoor environments often require many parameters, and complicated behaviors can be difficult or impossible to specify with a manageable number of parameters. Furthermore, their design requires knowledge of the robot's internal models and knowledge of the environment in which the behaviors will be used. In real-world scenarios, we can design new behaviors using our learning system much more quickly than we can write hand-crafted behaviors. We present the results of training the robot to execute several specialized and general-purpose behaviors, including traversing a slalom, staying near "cover," navigating on paths, navigating in an obstacle field, and general-purpose navigation. Our system learns and executes most of these behaviors well after 1-4 h of operator training time. In quantitative tests, the learned behavior is not as robust as a hand-crafted behavior but often completes obstacle courses more quickly. Additionally, we identify the factors that influence the effectiveness of this approach and investigate the properties of the training data provided by the human trainer. On the basis of our analyses, we suggest future work to ensure sufficient training, handle conflicting training examples, model robot dynamics, and further investigate dimensionality reduction of perception features. C 2009 Wiley Periodicals, Inc.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Learning outdoor mobile robot behaviors by example

Pippin¹,

Balch²

2009

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…The interaction between the two layers has been identified as crucial, and different approaches have been followed to realise it, leading to the development of three-layered architectures, in which a middle layer is dedicated to the interaction between the other two layers [6,32,79,104]. A three-layered control system which uses different strategies for robot control, starting with a high-frequency control loop for issuing motor commands at the lowest layer to a Dijkstra-based planner for long-range navigation is described in [95].…”

Section: Architectures Of Robot Control Systemsmentioning

confidence: 99%

Off-road Robotics—An Overview

Berns

Kuhnert²,

Armbrust

2011

Künstl Intell

View full text Add to dashboard Cite

show abstract

“…The classic separation results in a low-level system for local navigation that has a limited view on the world and operates in a fast, yet shortsighted manner. Different approaches have been followed to establish the interaction of the two parts ( (Wooden et al, 2007), (Ranganathan and Koenig, 2003)). An approach for detecting narrow passages in indoor environments is described in (Schröter, 2005).…”

Section: State Of the Artmentioning

confidence: 99%

Using Passages to Support Off-Road Robot Navigation

2009

Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics

View full text Add to dashboard Cite

Abstract:In this paper an approach for the detection of passages and their use in autonomous off-road robot navigation is presented. The authors argue that many two-layered architectures of robot navigation systems suffer from the gap between the typically coarse-grained high-level path-planning and the basically reactive low-level collision avoidance. In this context, passages shall be defined as paths leading through obstacles. The proposed approach is based on the idea that passages in the proximity to the robot should be evaluated with respect to their relevance for reaching the target area in order to avoid local detours by following suitable passages. The detection and assessment of passages is based on virtual sensors, a standardized data representation offering a unified, straightforward, and flexible retrieval mechanism for accessing the data provided by different sensor systems. For the evaluation of passages the authors introduce the concept of virtual sensor probes which can move independently from the robot. That way the point of view on the environment information can be tailored to support the detection and evaluation strategy. The proposed approach was deployed on the mobile off-road platform RAVON which serves as a testbed for the experiments carried out in the context of this work.

show abstract

Control-driven mapping and planning

Cited by 6 publications

References 13 publications

Learning outdoor mobile robot behaviors by example

Learning outdoor mobile robot behaviors by example

Off-road Robotics—An Overview

Using Passages to Support Off-Road Robot Navigation

Contact Info

Product

Resources

About