A representation method based on the probability of collision for safe robot navigation in domestic environments

Lunenburg, J. J. M.; Molengraft, René van de; Steinbuch, M Maarten

doi:10.1007/s10514-017-9653-x

“…Courtesy ScoutDI. consider a single, or a few, sources of risk (often called hazards), such as the variance in the current and future position of the robotic system and dynamic obstacles [3]- [5], variance in the future position of the robotic system and mapping uncertainty [6], [7], uncertainty in traffic density [8], or time, distance, and/or relative speed to obstacles [9]- [12]. These works demonstrate how risk information can increase the situation awareness of the system, but do not produce the holistic understanding needed to ensure safety.…”

Section: A Motivationmentioning

confidence: 99%

Supervisory Risk Control with Application to Industrial Drone Inspection

Rothmund¹,

Thieme²,

Utne³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

This article develops and experimentally tests a supervisory risk controller used to increase the safety of drone operations. Its task is to monitor the state of the drone and environment and to use this information to automatically change safety-critical parameters in real-time during operation. A case study of a tethered industrial inspection drone is considered. A system theoretic process analysis (STPA) is performed to identify how the system can fail. A Dynamic Decision Network (DDN), used as an online risk model, is built based on the results of the STPA. An optimization approach is used to choose an optimal parameter configuration that ensures an acceptable risk level. Through experimental tests, it is demonstrated how the supervisory risk controller is able to identify the state of the drone and the environment by combining information from multiple measurements over time and how it chooses values for the maximum speed, safety distance, and maximum vertical acceleration that produces an acceptable risk level. The parameters are updated during flight based on the output from the supervisory risk controller. When no parameter set can ensure an acceptable risk level then a recommendation of aborting the mission is sent to the human operator. Video of the experimental results can be found at https://youtu.be/RKhG9bguRJY

show abstract

“…Courtesy ScoutDI. consider a single, or a few, sources of risk (often called hazards), such as the variance in the current and future position of the robotic system and dynamic obstacles [3]- [5], variance in the future position of the robotic system and mapping uncertainty [6], [7], uncertainty in traffic density [8], or time, distance, and/or relative speed to obstacles [9]- [12]. These works demonstrate how risk information can increase the situation awareness of the system, but do not produce the holistic understanding needed to ensure safety.…”

Section: A Motivationmentioning

confidence: 99%

Supervisory Risk Control with Application to Industrial Drone Inspection

Rothmund¹,

Thieme²,

Utne³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

This article develops and experimentally tests a supervisory risk controller used to increase the safety of drone operations. Its task is to monitor the state of the drone and environment and to use this information to automatically change safety-critical parameters in real-time during operation. A case study of a tethered industrial inspection drone is considered. A system theoretic process analysis (STPA) is performed to identify how the system can fail. A Dynamic Decision Network (DDN), used as an online risk model, is built based on the results of the STPA. An optimization approach is used to choose an optimal parameter configuration that ensures an acceptable risk level. Through experimental tests, it is demonstrated how the supervisory risk controller is able to identify the state of the drone and the environment by combining information from multiple measurements over time and how it chooses values for the maximum speed, safety distance, and maximum vertical acceleration that produces an acceptable risk level. The parameters are updated during flight based on the output from the supervisory risk controller. When no parameter set can ensure an acceptable risk level then a recommendation of aborting the mission is sent to the human operator. Video of the experimental results can be found at https://youtu.be/RKhG9bguRJY

show abstract

“…Although this paper takes a deep RL approach to safe and low-risk robot navigation, there is a vast literature on classical model-predictive-control (MPC) and graph-search approaches. This literature considers diverse sources of risk, ranging from simple sensor noise and occlusion [13], [14], to uncertainty about the traversability of the edges (e.g. doors) of a navigation graph [15], and the unpredictability of pedestrian movements [16].…”

Section: R W a Risk In Mobile-robot Navigationmentioning

confidence: 99%

Risk-Conditioned Distributional Soft Actor-Critic for Risk-Sensitive Navigation

Choi¹,

Dance²,

Kim³

et al. 2021

Preprint

0

View full text Add to dashboard Cite

Modern navigation algorithms based on deep reinforcement learning (RL) show promising efficiency and robustness. However, most deep RL algorithms operate in a riskneutral manner, making no special attempt to shield users from relatively rare but serious outcomes, even if such shielding might cause little loss of performance. Furthermore, such algorithms typically make no provisions to ensure safety in the presence of inaccuracies in the models on which they were trained, beyond adding a cost-of-collision and some domain randomization while training, in spite of the formidable complexity of the environments in which they operate. In this paper, we present a novel distributional RL algorithm that not only learns an uncertainty-aware policy, but can also change its risk measure without expensive fine-tuning or retraining. Our method shows superior performance and safety over baselines in partiallyobserved navigation tasks. We also demonstrate that agents trained using our method can adapt their policies to a wide range of risk measures at run-time.

show abstract

A representation method based on the probability of collision for safe robot navigation in domestic environments

Cited by 8 publications

References 27 publications

Supervisory Risk Control with Application to Industrial Drone Inspection

Supervisory Risk Control with Application to Industrial Drone Inspection

Supervisory Risk Control with Application to Industrial Drone Inspection

Risk-Conditioned Distributional Soft Actor-Critic for Risk-Sensitive Navigation

Contact Info

Product

Resources

About