Guaranteed Safe Online Learning via Reachability: tracking a ground target using a quadrotor

By Jeremy H. Gillula, Claire Jennifer Tomlin. 2012

108 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Artificial Intelligence

Citations in the corpus

Uncertainty-Aware Reinforcement Learning for Collision Avoidance (2017, background): “There is much prior work on safe robot control for safety-critical systems such as autonomous cars [29], legged robots [31], and quadrotors [20, 30, 9].”
Safe Control under Uncertainty (2015, background): “More recently, safe learning approaches construct controllers that keep the system in the safe region, while the optimal strategy is learned online [15, 3, 1].”
Reachability-based safe learning with Gaussian processes (2014, methodology): “In the literature this has typically been achieved by using conservative bounds on the disturbance [8].”
Penalizing Side Effects using Stepwise Relative Reachability (2018, methodology): “…methods: methods that preserve reversibility (the reachability of a starting state) [Moldovan and Abbeel, 2012; Eysenbach et al., 2017], and reachability analysis methods that require reachability of a safe region [Mitchell et al., 2005; Gillula and Tomlin, 2012; Fisac et al., 2017].”
Measuring and avoiding side effects using relative reachability (2018, background): “A similar argument applies to reachability analysis methods (Mitchell et al., 2005; Gillula and Tomlin, 2012; Fisac et al., 2017), which require that a safe region must be reachable by a known conservative policy: the agent would be penalized if another agent or an environment event make the safe…”
Recovery RL: Safe Reinforcement Learning With Learned Recovery Zones (2020, methodology): “First, inspired by recent work in robust control [4, 11, 13, 21], we represent the RL agent with two policies: the ﬁrst policy focuses on optimizing the unconstrained task objective (task policy) and the second policy takes control when the task policy is in danger of constraint violations in the…”
Reducing Conservativeness in Safety Guarantees by Learning Disturbances Online: Iterated Guaranteed Safe Online Learning (2012, methodology): “For an easily understandable overview of HJI reachability, we refer the reader to Ding et al. [4].”
Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning (2021, methodology): “The general HJ reachability analysis framework (132) has also been combined with on-line dynamics model learning for a target tracking task (134), with online planning for safe exploration (135), and with temporal diﬀerence algorithms for safe RL (136).”
Safe Control under Uncertainty with Probabilistic Signal Temporal Logic (2016, background): “Designing controllers under reachability analysis and safe learning are well-studied methods that allow specifying safety and reachability properties, while learning the optimal strategy online [32, 33, 15, 3, 1].”
A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems (2017, background): “Initial work exploring this approach was presented in [26] and [27].”