Mapless Navigation among Dynamics with Social-safety-awareness: a reinforcement learning approach from 2D laser scans

Jin, Jun; Nguyen, Nhat M.; Sakib, Nazmus; Graves, Daniel; Yao, Hengshuai; Jägersand, Martin

doi:10.1109/icra40945.2020.9197148

Cited by 45 publications

(27 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… Katyal et al (2020) and Liu L. et al (2020) followed a similar method, including a simulation evaluation and a physical demonstration in their investigation of adaptive crowd navigation. Prior works that report this type of physical demonstration typically provide supplementary videos of the demonstrations (e.g., Jin et al, 2019 ).…”

Section: Evaluation Methods Scenarios and Datasetsmentioning

confidence: 99%

“…In general, prior works used navigation efficiency ( Guzzi et al, 2013a ; Guzzi et al, 2013b ; Mavrogiannis et al, 2018 ; Liang et al, 2020 ) and success rate ( Burgard et al, 1998 ; Guzzi et al, 2013b ; Jin et al, 2019 ; Liang et al, 2020 ; Nishimura and Yonetani, 2020 ; Tsai and Oh, 2020 ) to quantify the navigation performance of a robot. The common metrics for navigation performance are shown in Table 4 .…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…In addition to the efficiency metrics discussed above, success rate is commonly used to quantify navigation performance in socially-aware navigation ( Burgard et al, 1998 ; Guzzi et al, 2013b ; Jin et al, 2019 ; Liang et al, 2020 ; Nishimura and Yonetani, 2020 ; Tsai and Oh, 2020 ). Success rate, or arrival rate, measures an agent’s ability to reach its goal.…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…However, as we discussed earlier, additional representations of personal space have been proposed to capture nuanced social rules, cultural influences, and specific situations ( Rios-Martinez et al, 2015 ); therefore, this index may be adapted to account for different cultures, types of relationships, and interaction contexts by modifying the standard deviations. As another example of custom safety zones, Jin et al (2019) defined the ego-safety zone as a circular space around an agent, analogous to the personal space, and the social-safety zone as a rectangular region stretching along an agent’s current moving direction.…”

Section: Evaluation Metricsmentioning

confidence: 99%

See 3 more Smart Citations

Evaluation of Socially-Aware Robot Navigation

Gao

Huang

2022

Front. Robot. AI

View full text Add to dashboard Cite

As mobile robots are increasingly introduced into our daily lives, it grows ever more imperative that these robots navigate with and among people in a safe and socially acceptable manner, particularly in shared spaces. While research on enabling socially-aware robot navigation has expanded over the years, there are no agreed-upon evaluation protocols or benchmarks to allow for the systematic development and evaluation of socially-aware navigation. As an effort to aid more productive development and progress comparisons, in this paper we review the evaluation methods, scenarios, datasets, and metrics commonly used in previous socially-aware navigation research, discuss the limitations of existing evaluation protocols, and highlight research opportunities for advancing socially-aware robot navigation.

show abstract

Section: Evaluation Methods Scenarios and Datasetsmentioning

confidence: 99%

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Evaluation Metricsmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Socially-Aware Robot Navigation

Gao

Huang

2022

Front. Robot. AI

View full text Add to dashboard Cite

show abstract

“…• MNDS [4] is a costmap-based RL dynamic collision avoidance algorithm using laser scans in disentangled angle range representation as input. In the training procedure, as we can see in Fig.…”

Section: B Comparative Studymentioning

confidence: 99%

Learning Observation-Based Certifiable Safe Policy for Decentralized Multi-Robot Navigation

Cui¹,

Longzhong²,

Huang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Safety is of great importance in multi-robot navigation problems. In this paper, we propose a control barrier function (CBF) based optimizer that ensures robot safety with both high probability and flexibility, using only sensor measurement. The optimizer takes action commands from the policy network as initial values and then provides refinement to drive the potentially dangerous ones back into safe regions. With the help of a deep transition model that predicts the evolution of surrounding dynamics and the consequences of different actions, the CBF module can guide the optimization in a reasonable time horizon. We also present a novel joint training framework that improves the cooperation between the Reinforcement Learning (RL) based policy and the CBF-based optimizer both in training and inference procedures by utilizing reward feedback from the CBF module. We observe that the policy using our method can achieve a higher success rate while maintaining the safety of multiple robots in significantly fewer episodes compared with other methods. Experiments are conducted in multiple scenarios both in simulation and the real world, the results demonstrate the effectiveness of our method in maintaining the safety of multi-robot navigation.Code is available at https://github.com/YuxiangCui/ MARL-OCBF.

show abstract

Deep‐Q‐Network hybridization with extended Kalman filter for accelerate learning in autonomous navigation with auxiliary security module

Bezerra,

Vieira,

Soares

2024

Trans Emerging Tel Tech

View full text Add to dashboard Cite

This article proposes an algorithm for autonomous navigation of mobile robots that mixes reinforcement learning with extended Kalman filter (EKF) as a localization technique, namely EKF‐DQN, aiming to accelerate the maximization of the learning curve and improve the reward values obtained in the learning process. More specifically, Deep‐Q‐Networks (DQN) are used to control the trajectory of an autonomous robot in an environment with many obstacles. To improve navigation capability in this environment, we also propose a fusion of visual and nonvisual sensors. Due to the ability of EKF to predict states, this algorithm is used as a learning accelerator for the DQN network, predicting future states and inserting this information into the memory replay. Aiming to increase the safety of the navigation process, a visual safety system is also proposed to avoid collisions between the mobile robot and people circulating in the environment. The efficiency of the proposed control system is verified through computational simulations using the CoppeliaSIM simulator with code insertion in Python. The simulation results show that the EKF‐DQN algorithm accelerates the maximization of rewards obtained and provides a higher success rate in fulfilling the mission assigned to the robot when compared to other value‐based and policy‐based algorithms. A demo video of the navigation system can be seen at: https://bit.ly/3reEZrU.

show abstract

Mapless Navigation among Dynamics with Social-safety-awareness: a reinforcement learning approach from 2D laser scans

Cited by 45 publications

References 17 publications

Evaluation of Socially-Aware Robot Navigation

Evaluation of Socially-Aware Robot Navigation

Learning Observation-Based Certifiable Safe Policy for Decentralized Multi-Robot Navigation

Deep‐Q‐Network hybridization with extended Kalman filter for accelerate learning in autonomous navigation with auxiliary security module

Contact Info

Product

Resources

About