Towards Autonomous Grading In The Real World

Miron, Yakov; Ross, Chana; Goldfracht, Yuval; Tessler, Chen; Castro, Dotan Di

doi:10.48550/arxiv.2206.06091

Cited by 1 publication

(2 citation statements)

References 19 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This figure compares the performance of three agents: (i) A heuristic agent presented in [22], represented by a round marker, for comparison. (ii) agent 1 from [24], represented by a square marker. (iii) agent 2 , represented by a star marker.…”

Section: Results Analysismentioning

confidence: 99%

“…[22] presented a ML approach for autonomous grading that exhibited better generalization capabilities in previously unseen scenarios, both in simulation and in real-world settings. Moreover, [24] trained a privileged agent to mitigate the sim-to-real perception gap using a simulator. [25], developed a pioneering software package on a construction vehicle for automation; their excavator was one of the first to operate autonomously.…”

Section: Autonomous Construction Vehiclesmentioning

confidence: 99%

See 1 more Smart Citation

AG2U -- Autonomous Grading Under Uncertainties

Miron¹,

Goldfracht²,

Ross³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Surface grading, the process of leveling an uneven area containing pre-dumped sand piles, is an important task in the construction site pipeline. This labour-intensive process is often carried out by a dozer, a key machinery tool at any construction site. Current attempts to automate surface grading assume perfect localization. However, in real-world scenarios, this assumption fails, as agents are presented with imperfect perception, which leads to degraded performance. In this work, we address the problem of autonomous grading under uncertainties. First, we implement a simulation and a scaled real-world prototype environment to enable rapid policy exploration and evaluation in this setting. Second, we formalize the problem as a partially observable markov decision process and train an agent capable of handling such uncertainties. We show, through rigorous experiments, that an agent trained under perfect localization will suffer degraded performance when presented with localization uncertainties. However, an agent trained using our method will develop a more robust policy for addressing such errors and, consequently, exhibit a better grading performance.

show abstract

Section: Results Analysismentioning

confidence: 99%