Energy-Based Legged Robots Terrain Traversability Modeling via Deep Inverse Reinforcement Learning

Grizzle, Jessy W.; Eustice, Ryan M.; Ghaffari, Maani

doi:10.1109/lra.2022.3188100

Cited by 26 publications

(13 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future work includes investigating methods to compress and streamline data acquisition, working on data sets in unstructured environments and investigating memory-based alternatives to the autoregressive model. Exploiting the uncertainty estimated by D-BKI in scene understanding and robot navigation tasks [71] is also an attractive future work direction.…”

Section: Discussionmentioning

confidence: 99%

Dynamic Semantic Occupancy Mapping Using 3D Scene Flow and Closed-Form Bayesian Inference

et al. 2022

Self Cite

View full text Add to dashboard Cite

This paper reports on a dynamic semantic mapping framework that incorporates 3D scene flow measurements into a closed-form Bayesian inference model. Existence of dynamic objects in the environment can cause artifacts and traces in current mapping algorithms, leading to an inconsistent map posterior. We leverage state-of-the-art semantic segmentation and 3D flow estimation using deep learning to provide measurements for map inference. We develop a Bayesian model that propagates the scene with flow and infers a 3D continuous (i.e., can be queried at arbitrary resolution) semantic occupancy map outperforming its static counterpart. Extensive experiments using publicly available data sets show that the proposed framework improves over its predecessors and input measurements from deep neural networks consistently.

show abstract

Section: Discussionmentioning

confidence: 99%

Dynamic Semantic Occupancy Mapping Using 3D Scene Flow and Closed-Form Bayesian Inference

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition, the IIG algorithm (Ghaffari Jadidi et al, 2019), combined with an MPC (Teng et al, 2021a), can provide an integrated kinodynamic planner that takes the robot stability, control constraints, and the value of information from sensory data into account. Gan et al (2022) show that the value of information can be learned from multimodal sensory input via learning from demonstrations and self-supervised trajectory ranking to deal with sub-optimal demonstrations. In Section 5, we showed how equivariant neural networks can serve as powerful feature learners to improve data efficiency and generalizability across different tasks.…”

Section: Figure 16mentioning

confidence: 99%

Progress in symmetry preserving robot perception and control through geometry and learning

et al. 2022

Self Cite

View full text Add to dashboard Cite

This article reports on recent progress in robot perception and control methods developed by taking the symmetry of the problem into account. Inspired by existing mathematical tools for studying the symmetry structures of geometric spaces, geometric sensor registration, state estimator, and control methods provide indispensable insights into the problem formulations and generalization of robotics algorithms to challenging unknown environments. When combined with computational methods for learning hard-to-measure quantities, symmetry-preserving methods unleash tremendous performance. The article supports this claim by showcasing experimental results of robot perception, state estimation, and control in real-world scenarios.

show abstract

“…However, the IRL-based methods highly rely on the quality of human demonstrations, resulting in bad navigation performance if the latter are not optimal. [13] proposes an energy-based trajectory ranking loss in the training process of MEDIRL to eliminate the effect of non-optimal and sub-optimal demonstrations. For our proposed method, we extend the usage of T-MEDIRL to the socially-aware navigation field using the sudden velocity change of nearby people as the trajectory ranking loss.…”

Section: B Irl-based Navigationmentioning

confidence: 99%

“…Hence, it highly relies on the quality of demonstrations. Inspired by [13], we add a trajectory loss term to the MEDIRL framework to solve this problem. As shown in Fig.…”

Section: B T-medirlmentioning

confidence: 99%

See 1 more Smart Citation

SoLo T-DIRL: Socially-Aware Dynamic Local Planner based on Trajectory-Ranked Deep Inverse Reinforcement Learning

Xu¹,

Chakhachiro²,

Kathuria³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This work proposes a new framework for a socially-aware dynamic local planner in crowded environments by building on the recently proposed Trajectory-ranked Maximum Entropy Deep Inverse Reinforcement Learning (T-MEDIRL). To address the social navigation problem, our multi-modal learning planner explicitly considers social interaction factors, as well as social-awareness factors into T-MEDIRL pipeline to learn a reward function from human demonstrations. Moreover, we propose a novel trajectory ranking score using the sudden velocity change of pedestrians around the robot to address the sub-optimality in human demonstrations.Our evaluation shows that this method can successfully make a robot navigate in a crowded social environment and outperforms the state-of-art social navigation methods in terms of the success rate, navigation time, and invasion rate.

show abstract

Energy-Based Legged Robots Terrain Traversability Modeling via Deep Inverse Reinforcement Learning

Cited by 26 publications

References 33 publications

Dynamic Semantic Occupancy Mapping Using 3D Scene Flow and Closed-Form Bayesian Inference

Dynamic Semantic Occupancy Mapping Using 3D Scene Flow and Closed-Form Bayesian Inference

Progress in symmetry preserving robot perception and control through geometry and learning

SoLo T-DIRL: Socially-Aware Dynamic Local Planner based on Trajectory-Ranked Deep Inverse Reinforcement Learning

Contact Info

Product

Resources

About