Extending the Knowledge of Volumes approach to robot task planning with efficient geometric predicates

Gaschler, Andre; Kessler, Ingmar; Petrick, Ronald P. A.; Knoll, Alois

doi:10.1109/icra.2015.7139619

Cited by 14 publications

(9 citation statements)

References 14 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Altogether it features 15 degrees-of-freedom, but we will only plan for the torso and the arm, resulting in a 10-dimensional configuration space for the robot. While we show a detailed CAD model for visualization purposes, our collision detection routine utilizes a bounding convex decomposition for efficiency [17], [18]. In this scenario, the robot acts as a bartender and is supposed to clean up the workspace by aligning all three bottles on the lower support surface (Figure 1a).…”

Section: Discussionmentioning

confidence: 99%

MOPL: A multi-modal path planner for generic manipulation tasks

Jentzsch

Gaschler

Khatib

et al. 2015

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

View full text Add to dashboard Cite

Abstract-For intelligent robots to solve real-world tasks, they need to manipulate multiple objects, and perform diverse manipulation actions apart from rigid transfers, such as pushing and sliding. Planning these tasks requires discrete changes between actions, and continuous, collision-free paths that fulfill action-specific constraints. In this work, we propose a multi-modal path planner, named MOPL, which accepts generic definitions of primitive actions with different types of contact manifolds, and randomly spans its search trees through these subspaces. Our evaluation shows that this generic search technique allows MOPL to solve several challenging scenarios over different types of kinematics and tools with reasonable performance. Furthermore, we demonstrate MOPL by solving and executing plans in two real-world experimental setups.

show abstract

Section: Discussionmentioning

confidence: 99%

MOPL: A multi-modal path planner for generic manipulation tasks

Jentzsch

Gaschler

Khatib

et al. 2015

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Rather than following greedy heuristics that cannot solve every case, a combined task and motion planner was developed to search in the full space of discrete actions and motion paths to solve generic pick-and-place tasks [26,27]. The integrated planner can start from an arbitrary world state, such as one where random bottles have been placed on a bar table by human interaction partners.…”

Section: Task and Motion Planning For Interactive Manipulationmentioning

confidence: 99%

“…As an example, when a customer has placed an empty bottle on the right side of the bar table, only one arm can pick it up, but it needs to put it down in the common workspace where the other arm can reach it to transfer it to the empty bottle storage region [27]. To allow concise definition of scenarios and goal conditions, predicates to model support surfaces and the inclusion relation are also available [26]. The search takes a few seconds for scenarios with four objects and two arms, with most of the time spent on collision checking.…”

Section: Task and Motion Planning For Interactive Manipulationmentioning

confidence: 99%

Applications in HHI: Physical Cooperation

Rickert

Gaschler

Knoll

2018

Humanoid Robotics: A Reference

Self Cite

View full text Add to dashboard Cite

Humans critically depend on permanent verbal and nonverbal interaction-for aligning their mental states, for synchronizing their intentions and goals, and also for performing joint tasks, such as carrying a heavy object together, manipulating of objects in a common workspace, or handing over components and building or assembling larger structures in teams. Typically, physical interaction is initiated by a short joint planning dialog and then further accompanied by a stream of verbal utterances. For obtaining a smooth interaction flow in a given situation, humans typically use all their communication modalities and senses, and this often happens even unconsciously. As we move toward the introduction of robotic co-workers that serve humans-some of them will be humanoids; others will be of a different shape-humans will expect them to be integrated into the

show abstract

“…Path planning algorithms are essential for the accomplishment of many activities in different areas, for example, robot navigation [1], path apps for locomotion in cities (for pedestrian and driver) [2], autonomous-driver cars [3]. These algorithms have different approaches to treat spatial information, the most used in the literature are, Grid-based search (which transforms the environment in a grid-mesh) [4], Intervalbased search (similar to grid-based search it but uses space data instead of a grid) [5] and Reward-based (similar to a reinforcement learning in deep learning) [6].…”

Section: Introductionmentioning

confidence: 99%

Performance Improvement of Path Planning algorithms with Deep Learning Encoder Model

Ferreira¹,

Júnior²,

Galvão³

et al. 2020

Preprint

View full text Add to dashboard Cite

Currently, path planning algorithms are used in many daily tasks. They are relevant to find the best route in traffic and make autonomous robots able to navigate. The use of path planning presents some issues in large and dynamic environments. Large environments make these algorithms spend much time finding the shortest path. On the other hand, dynamic environments request a new execution of the algorithm each time a change occurs in the environment, and it increases the execution time. The dimensionality reduction appears as a solution to this problem, which in this context means removing useless paths present in those environments. Most of the algorithms that reduce dimensionality are limited to the linear correlation of the input data. Recently, a Convolutional Neural Network (CNN) Encoder was used to overcome this situation since it can use both linear and non-linear information to data reduction. This paper analyzes in-depth the performance to eliminate the useless paths using this CNN Encoder model. To measure the mentioned model efficiency, we combined it with different path planning algorithms. Next, the final algorithms (combined and not combined) are checked in a database that is composed of five scenarios. Each scenario contains fixed and dynamic obstacles. Their proposed model, the CNN Encoder, associated to other existent path planning algorithms in the literature, was able to obtain a time decrease to find the shortest path in comparison to all path planning algorithms analyzed. the average decreased time was 54.43%.

show abstract

Extending the Knowledge of Volumes approach to robot task planning with efficient geometric predicates

Cited by 14 publications

References 14 publications

MOPL: A multi-modal path planner for generic manipulation tasks

MOPL: A multi-modal path planner for generic manipulation tasks

Applications in HHI: Physical Cooperation

Performance Improvement of Path Planning algorithms with Deep Learning Encoder Model

Contact Info

Product

Resources

About