2018
DOI: 10.1177/0278364918784660
|View full text |Cite
|
Sign up to set email alerts
|

Effective metrics for multi-robot motion-planning

Abstract: We study the effectiveness of metrics for Multi-Robot Motion-Planning (MRMP) when using RRT-style sampling-based planners. These metrics play the crucial role of determining the nearest neighbors of configurations and in that they regulate the connectivity of the underlying roadmaps produced by the planners and other properties like the quality of solution paths. After screening over a dozen different metrics we focus on the five most promising ones-two more traditional metrics, and three novel ones which we p… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
6
1

Relationship

1
6

Authors

Journals

citations
Cited by 12 publications
(8 citation statements)
references
References 75 publications
0
8
0
Order By: Relevance
“…Neural network training aims to minimize the error function E, which is the mean squared error (MSE) that quantifies the difference between the computed output trajectory (feedforward neural network) and the actual trajectory (mathematical model). The number of layers and parameters were determined empirically by examining the performance of several different layers (3)(4)(5)(6)(7)(8)(9)(10)(11)(12) and parameters . We found that 100 parameters per layer and 4 total layers gave the least mean squared error (M SE < 0.001).…”
Section: Supervised Learning Approachmentioning
confidence: 99%
See 1 more Smart Citation
“…Neural network training aims to minimize the error function E, which is the mean squared error (MSE) that quantifies the difference between the computed output trajectory (feedforward neural network) and the actual trajectory (mathematical model). The number of layers and parameters were determined empirically by examining the performance of several different layers (3)(4)(5)(6)(7)(8)(9)(10)(11)(12) and parameters . We found that 100 parameters per layer and 4 total layers gave the least mean squared error (M SE < 0.001).…”
Section: Supervised Learning Approachmentioning
confidence: 99%
“…As with swarm trajectory planning, collision-free paths must be generated, which is well-known to be NP-hard [11]. Many works such as [12], [13], [14], [15] have explored multi-robot path planning in more general contexts.…”
Section: Introductionmentioning
confidence: 99%
“…Later, dRRT was applied to motion planning of a free-flying multi-link robot (Salzman et al, 2016). In that case, dRRT allowed to efficiently decouple between costly self-collision checks, which were done offline, and robot-obstacle collision checks, by traversing an implicitly-defined roadmap, whose structure resembles to that of Ĝ. dRRT has also been used in the study of the effectiveness of metrics for MMP, which are an essential ingredient in sampling-based planners (Atias et al, 2017).…”
Section: Prior Workmentioning
confidence: 99%
“…Several sampling-based motion planning algorithms have been proposed for multi-agent systems [348]- [351]; in particular, [349], [351] offer optimality guarantees. A key problem in sampling-based motion planning is to efficiently assess the "distance" between sampled points -the work in [352] proposes and assesses the performance of different metrics to efficiently estimate the "distance" between samples in the multi-agent setting.…”
Section: E Coordinated Combinatorial and Sampling-based Motion-planning Algorithmsmentioning
confidence: 99%