Motion Planning and Control for Mobile Robot Navigation Using Machine Learning: a Survey

Xiao, Xuesu; Liu, Bo; Warnell, Garrett; Stone, Peter

doi:10.48550/arxiv.2011.13112

Cited by 9 publications

(16 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning approaches have been applied to the classical navigation pipeline in different ways [1], such as constructing a world representation [4], [5], fine-tuning planner parameters [6]- [8], improving navigation performance with experience [2], or enabling social [17] and terrainaware navigation [3]. Most learning methods require either extensive (RL) or high-quality (IL) training data, such as that derived from trial-and-error exploration or from human demonstrations, respectively.…”

Section: B Machine Learning For Navigationmentioning

confidence: 99%

“…Although classical navigation systems can safely and reliably move mobile robots from one point to another within obstacle-occupied environments, recent machine learning techniques have demonstrated improvement over their classical counterparts [1], e.g., by learning local planners [2], [3], learning world representation [4], [5], or learning planner parameters [6]- [8]. However, these learning approaches heavily depend on access to high quality training data.…”

Section: Introductionmentioning

confidence: 99%

“…We represent the random exploration policy π rand to collect p in an open space using Ego-Planner with 2.0m/s max v in a highly constrained environment. We only record motion plans p but do not record any perception input (as if the UAV were flying in an open space),1 which is later synthesized by LfLH. Note that previous learning approaches from hallucination cannot solve this 3D aerial navigation task: LfH[9] is not applicable because no global path is available; The approximated minimal unreachable set in HLSD[10] cannot effectively represent 3D obstacles given the long flight plan.For our LfLH implementation in 3D, the odometry point contains the same information (x i , φ i , v i , ω i ) as the UGV (but in 3D), and each plan p consists of M = 500 odometry points that takes 2.5s.…”

mentioning

confidence: 99%

See 2 more Smart Citations

From Agile Ground to Aerial Navigation: Learning from Learned Hallucination

Wang¹,

Xiao²,

Nettekoven³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents a self-supervised Learning from Learned Hallucination (LfLH) method to learn fast and reactive motion planners for ground and aerial robots to navigate through highly constrained environments. The recent Learning from Hallucination (LfH) paradigm for autonomous navigation executes motion plans by random exploration in completely safe obstacle-free spaces, uses hand-crafted hallucination techniques to add imaginary obstacles to the robot's perception, and then learns motion planners to navigate in realistic, highly-constrained, dangerous spaces. However, current handcrafted hallucination techniques need to be tailored for specific robot types (e.g., a differential drive ground vehicle), and use approximations heavily dependent on certain assumptions (e.g., a short planning horizon). In this work, instead of manually designing hallucination functions, LfLH learns to hallucinate obstacle configurations, where the motion plans from random exploration in open space are optimal, in a self-supervised manner. LfLH is robust to different robot types and does not make assumptions about the planning horizon. Evaluated in both simulated and physical environments with a ground and an aerial robot, LfLH outperforms or performs comparably to previous hallucination approaches, along with sampling-and optimization-based classical methods.

show abstract

Section: B Machine Learning For Navigationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

From Agile Ground to Aerial Navigation: Learning from Learned Hallucination

Wang¹,

Xiao²,

Nettekoven³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…While autonomous navigation has been studied by the robotics community for decades, machine learning approaches have recently been extensively applied to this problem as well. Xiao, et al [11] presented a survey on using machine learning for motion control in mobile robot navigation: while the majority of learning approaches tackle navigation in an end-to-end manner [12], [13], it was found that approaches using learning in conjunction with other classical navigation components were more likely to have achieved better navigation performance. These methods included those that learned sub-goals [14], local planners [6]- [8], [15], [16], or planner parameters [4], [5], [9], [17]- [19].…”

Section: A Learning For Navigationmentioning

confidence: 99%

APPLE: Adaptive Planner Parameter Learning from Evaluative Feedback

Wang,

Xiao,

Warnell

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Classical autonomous navigation systems can control robots in a collision-free manner, oftentimes with verifiable safety and explainability. When facing new environments, however, finetuning of the system parameters by an expert is typically required before the system can navigate as expected. To alleviate this requirement, the recently-proposed Adaptive Planner Parameter Learning paradigm allows robots to learn how to dynamically adjust planner parameters using a teleoperated demonstration or corrective interventions from non-expert users. However, these interaction modalities require users to take full control of the moving robot, which requires the users to be familiar with robot teleoperation. As an alternative, we introduce APPLE, Adaptive Planner Parameter Learning from Evaluative Feedback (realtime, scalar-valued assessments of behavior), which represents a less-demanding modality of interaction. Simulated and physical experiments show APPLE can achieve better performance compared to the planner with static default parameters and even yield improvement over learned parameters from richer interaction modalities. I. INTRODUCTIONM OBILE robot navigation is a well-studied problem in the robotics community. Many classical approaches have been developed over the last several decades and several of them have been robustly deployed on physical robot platforms moving in the real world [1], [2], with verifiable guarantees of safety and explainability.However, prior to deployment in a new environment, these approaches typically require parameter re-tuning in order to achieve robust navigation performance. For example, in cluttered environments, a low velocity and high sampling rate are necessary in order for the system to be able to generate safe and smooth motions, whereas in relatively open spaces, a

show abstract

“…With the recent advances in machine learning research, data-driven techniques have started being applied to the mobile robot navigation problem. Xiao, et al [6] presented a survey on using machine learning for motion control in mobile robot navigation: in addition to some works in the emerging social [7] and terrain-based [8] navigation, most existing learning approaches for navigation adopt an end-toend learning paradigm to address the classical collision-free navigation problem and are capable of generating navigational behaviors [9], [10]. However, those end-to-end approaches still cannot outperform classical navigation methods or are not even compared to them.…”

Section: Arxiv:210507620v1 [Csro] 17 May 2021mentioning

confidence: 99%

APPLD: Adaptive Planner Parameter Learning From Demonstration

Xiao

Liu

Warnell

et al. 2020

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

While current autonomous navigation systems allow robots to successfully drive themselves from one point to another in specific environments, they typically require extensive manual parameter re-tuning by human robotics experts in order to function in new environments. Furthermore, even for just one complex environment, a single set of fine-tuned parameters may not work well in different regions of that environment. These problems prohibit reliable mobile robot deployment by nonexpert users. As a remedy, we propose Adaptive Planner Parameter Learning (APPL), a machine learning framework that can leverage non-expert human interaction via several modalitiesincluding teleoperated demonstrations, corrective interventions, and evaluative feedback-and also unsupervised reinforcement learning to learn a parameter policy that can dynamically adjust the parameters of classical navigation systems in response to changes in the environment. APPL inherits safety and explainability from classical navigation systems while also enjoying the benefits of machine learning, i.e., the ability to adapt and improve from experience. We present a suite of individual APPL methods and also a unifying cycle-of-learning scheme that combines all the proposed methods in a framework that can improve navigation performance through continual, iterative human interaction and simulation training.

show abstract

Motion Planning and Control for Mobile Robot Navigation Using Machine Learning: a Survey

Cited by 9 publications

References 83 publications

From Agile Ground to Aerial Navigation: Learning from Learned Hallucination

From Agile Ground to Aerial Navigation: Learning from Learned Hallucination

APPLE: Adaptive Planner Parameter Learning from Evaluative Feedback

APPLD: Adaptive Planner Parameter Learning From Demonstration

Contact Info

Product

Resources

About