Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Safe Autonomy for Uncrewed Surface Vehicles Using Adaptive Control and Reachability Analysis

Karan Mahesh, Tyler M. Paine, Max L. Greene, Nicholas Rober, Steven Lee, Sildomar T. Monteiro, Anuradha Annaswamy, Michael R. Benjamin, Jonathan P. How

Abstract

Marine robots must maintain precise control and ensure safety during tasks like ocean monitoring, even when encountering unpredictable disturbances that affect performance. Designing algorithms for uncrewed surface vehicles (USVs) requires accounting for these disturbances to control the vehicle and ensure it avoids obstacles. While adaptive control has addressed USV control challenges, real-world applications are limited, and certifying USV safety amidst unexpected disturbances remains difficult. To tackle control issues, we employ a model reference adaptive controller (MRAC) to stabilize the USV along a desired trajectory. For safety certification, we developed a reachability module with a moving horizon estimator (MHE) to estimate disturbances affecting the USV. This estimate is propagated through a forward reachable set calculation, predicting future states and enabling real-time safety certification. We tested our safe autonomy pipeline on a Clearpath Heron USV in the Charles River, near MIT. Our experiments demonstrated that the USV's MRAC controller and reachability module could adapt to disturbances like thruster failures and drag forces. The MRAC controller outperformed a PID baseline, showing a 45%-81% reduction in RMSE position error. Additionally, the reachability module provided real-time safety certification, ensuring the USV's safety. We further validated our pipeline's effectiveness in underway replenishment and canal scenarios, simulating relevant marine tasks.

Safe Autonomy for Uncrewed Surface Vehicles Using Adaptive Control and Reachability Analysis

Abstract

Marine robots must maintain precise control and ensure safety during tasks like ocean monitoring, even when encountering unpredictable disturbances that affect performance. Designing algorithms for uncrewed surface vehicles (USVs) requires accounting for these disturbances to control the vehicle and ensure it avoids obstacles. While adaptive control has addressed USV control challenges, real-world applications are limited, and certifying USV safety amidst unexpected disturbances remains difficult. To tackle control issues, we employ a model reference adaptive controller (MRAC) to stabilize the USV along a desired trajectory. For safety certification, we developed a reachability module with a moving horizon estimator (MHE) to estimate disturbances affecting the USV. This estimate is propagated through a forward reachable set calculation, predicting future states and enabling real-time safety certification. We tested our safe autonomy pipeline on a Clearpath Heron USV in the Charles River, near MIT. Our experiments demonstrated that the USV's MRAC controller and reachability module could adapt to disturbances like thruster failures and drag forces. The MRAC controller outperformed a PID baseline, showing a 45%-81% reduction in RMSE position error. Additionally, the reachability module provided real-time safety certification, ensuring the USV's safety. We further validated our pipeline's effectiveness in underway replenishment and canal scenarios, simulating relevant marine tasks.
Paper Structure (58 sections, 2 theorems, 47 equations, 23 figures, 6 tables, 1 algorithm)

This paper contains 58 sections, 2 theorems, 47 equations, 23 figures, 6 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Adaptive Control in Marine Vehicles.
  4. Forward Reachability Analysis and Safety Assurances
  5. Model of USV Dynamics and Kinematics
  6. Linearized Models
  7. Linearized Speed Model
  8. Linearized Sway Yaw-Rate Model
  9. Discussion
  10. Discrete Closed-Loop Model
  11. Control Architecture
  12. Guidance Behavior Design
  13. Baseline Track-line Following
  14. L1 Guidance
  15. Underway Replenishment (UNREP)
  16. ...and 43 more sections

Key Result

Theorem 5.1

Given a CG $\bm{G}$ and a hyper-rectangular set of possible inputs $\mathcal{I}$, there exist two explicit functions such that the inequality $g^{\bm{G}}_{L,o}(\bm{z}) \leq g^{\bm{G}}_o(\bm{z}) \leq g^{\bm{G}}_{U,o}(\bm{z})$ holds element-wise for all $\bm{z} \in \mathcal{I}$, with $\bm{\Psi}, \bm{\Phi} \in \mathbb{R}^{n_o \times n_i}$ and $\bm{\alpha}, \bm{\beta} \in \mathbb{R}^{n_o}$.

Figures (23)

  • Figure 1: A diagram showing our approach to control and safety certification applied to a Clearpath Heron USV. The goal of the controller is to overcome disturbances and keep the trajectory with disturbances close to the nominal trajectory. The goal of reachability analysis is to generate reachable sets that bound the true (i.e., disturbed) trajectory.
  • Figure 2: The block diagram of the control and reachability analysis architecture. The reference model is used in both the MRAC approach and the forward reachability analysis with the MHE.
  • Figure 3: Clearpath Robotics Heron USV operated on the Charles River.
  • Figure 4: Block diagram depicting our approach rober2023online. Data is collected from the system and fed into the MHE, which estimates the state and disturbance terms. The outputs of the MHE are used with a CG representation of the closed-loop dynamics to conduct reachability analysis and certify safety.
  • Figure 5: System architecture detailing backseat payload autonomy with commercial Heron frontseat.
  • ...and 18 more figures

Theorems & Definitions (2)

  • Theorem 5.1: Linear Relaxation of CGs, Thm. 2 xu2020automatic
  • Theorem 5.2: Safety Verification for an Uncertain System