HouseExpo: A Large-scale 2D Indoor Layout Dataset for Learning-based Algorithms on Mobile Robots

Li, Tingguang; Ho, Danny; Li, Chenming; Zhu, Delong; Wang, Chaoqun; Meng, Max Q.-H.

doi:10.1109/iros45743.2020.9341284

Cited by 32 publications

(22 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) HouseExpo Benchmarking: This benchmarking involves the generalized models running on maps from the HouseExpo dataset [43], modified to a 100×100 size.…”

Section: Resultsmentioning

confidence: 99%

“…We compare WPN with VINs, as well as MPNet [14] and two other RNN-based algorithms [39] and [25]. The comparisons are preformed on synthetic maps of different sizes [12], realworld maps [40], [41], [42], and HouseExpo maps [43]. The view module utilizes the current information provided by the sensor, in this case a typical measurement from a scanning laser ranger.…”

Section: Related Workmentioning

confidence: 99%

“…This experiment is designed to demonstrate the transferablity of the learned algorithm from synthetic maps to realworld maps. Here we use HouseExpo dataset [43], which is a large-scale 2D floor plan dataset built on SUNCG [47] dataset. We also test WPN on occupancy grid maps produced by SLAM.…”

Section: Houseexpo Benchmarking: Simulated Real-world Mapsmentioning

confidence: 99%

See 2 more Smart Citations

Waypoint Planning Networks

Toma

Jaafar

Hsueh

et al. 2021

Preprint

View full text Add to dashboard Cite

With the recent advances in machine learning, path planning algorithms are also evolving; however, the learned path planning algorithms often have difficulty competing with success rates of classic algorithms. We propose waypoint planning networks (WPN), a hybrid algorithm based on LSTMs with a local kernel-a classic algorithm such as A*, and a global kernel using a learned algorithm. WPN produces a more computationally efficient and robust solution. We compare WPN against A*, as well as related works including motion planning networks (MPNet) and value iteration networks (VIN). In this paper, the design and experiments have been conducted for 2D environments. Experimental results outline the benefits of WPN, both in efficiency and generalization. It is shown that WPN's search space is considerably less than A*, while being able to generate near optimal results. Additionally, WPN works on partial maps, unlike A* which needs the full map in advance. The code is available online 1 .

show abstract

“…2) HouseExpo Benchmarking: This benchmarking involves the generalized models running on maps from the HouseExpo dataset [43], modified to a 100×100 size.…”

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Waypoint Planning Networks

Toma

Jaafar

Hsueh

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The map prediction and motion planning networks were trained on two distinct datasets, dubbed D 1 and D 2 . Dataset D 1 holds 50, 000 independently generated maps, and D 2 contains 15, 894 maps from the HouseExpo dataset in [16]. Several examples from each dataset are displayed in Figure 3.…”

Section: B Training Setsmentioning

confidence: 99%

Integrating Deep Reinforcement and Supervised Learning to Expedite Indoor Mapping

Zwecher¹,

Iceland²,

R.³

et al. 2021

Preprint

View full text Add to dashboard Cite

The challenge of mapping indoor environments is addressed. Typical heuristic algorithms for solving the motion planning problem are frontier-based methods, that are especially effective when the environment is completely unknown. However, in cases where prior statistical data on the environment's architectonic features is available, such algorithms can be far from optimal. Furthermore, their calculation time may increase substantially as more areas are exposed. In this paper we propose two means by which to overcome these shortcomings. One is the use of deep reinforcement learning to train the motion planner. The second is the inclusion of a pre-trained generative deep neural network, acting as a map predictor. Each one helps to improve the decision making through use of the learned structural statistics of the environment, and both, being realized as neural networks, ensure a constant calculation time. We show that combining the two methods can shorten the mapping time, compared to frontier-based motion planning, by up to 75%.

show abstract

“…External maps can be imported into PathBench to diversify the datasets. Houseexpo [45] is a large dataset of 2D floor plans built on SUNCG dataset [46]. It contains 35,126 2D floor plans that have 252,550 rooms in total and can be used for PathBench benchmarking.…”

Section: External Mapsmentioning

confidence: 99%

PathBench: A Benchmarking Platform for Classical and Learned Path Planning Algorithms

Toma

Hsueh

Jaafar

et al. 2021

Preprint

View full text Add to dashboard Cite

Path planning is a key component in mobile robotics. A wide range of path planning algorithms exist, but few attempts have been made to benchmark the algorithms holistically or unify their interface. Moreover, with the recent advances in deep neural networks, there is an urgent need to facilitate the development and benchmarking of such learning-based planning algorithms. This paper presents PathBench, a platform for developing, visualizing, training, testing, and benchmarking of existing and future, classical and learned 2D and 3D path planning algorithms, while offering support for Robot Operating System (ROS). Many existing path planning algorithms are supported; e.g. A*, wavefront, rapidly-exploring random tree, value iteration networks, gated path planning networks; and integrating new algorithms is easy and clearly specified. We demonstrate the benchmarking capability of PathBench by comparing implemented classical and learned algorithms for metrics, such as path length, success rate, computational time and path deviation. These evaluations are done on built-in PathBench maps and external path planning environments from video games and real world databases. PathBench is open source 1 .

show abstract

HouseExpo: A Large-scale 2D Indoor Layout Dataset for Learning-based Algorithms on Mobile Robots

Cited by 32 publications

References 26 publications

Waypoint Planning Networks

Waypoint Planning Networks

Integrating Deep Reinforcement and Supervised Learning to Expedite Indoor Mapping

PathBench: A Benchmarking Platform for Classical and Learned Path Planning Algorithms

Contact Info

Product

Resources

About