Bionic Sensing with Artificial Lateral Line Systems for Fish-Like Underwater Robots

Cover
Half Title
Title Page
Copyright Page
Contents
CHAPTER 1: Introduction
	1.1. RESEARCH BACKGROUND
	1.2. DOCUMENT STRUCTURE
CHAPTER 2: Fish Lateral Line Inspired Perception and Flow-Aided Control: A Review
	2.1. INTRODUCTION
	2.2. MECHANISMS AND MODELS OF THE FISH LATERAL LINE
	2.3. THE EXISTING ALL SENSORS AND SYSTEMS
		2.3.1. ALL Sensor Unit
			2.3.1.1. Piezoresistive ALL Sensors
			2.3.1.2. Piezoelectric ALL Sensors
			2.3.1.3. Capacitive ALL Sensors
			2.3.1.4. Optical ALL Sensors
			2.3.1.5. Hot-Wire ALL Sensors
		2.3.2. ALL Sensors Placement Optimization
	2.4. HYDRODYNAMIC ENVIRONMENT SENSING AND VORTICES DETECTION
		2.4.1. Flow Field Characteristics Identification
		2.4.2. Flow Velocity and Direction Detection
		2.4.3. Vortex Street Properties Detection
	2.5. ALL-BASED DIPOLE SOURCE DETECTION
	2.6. FLOW-AIDED CONTROL OF UNDERWATER ROBOTS USING ALL SYSTEM
		2.6.1. Pattern Identification
		2.6.2. Motion Parameters (Speed and Direction) Estimation and Control
		2.6.3. Obstacles Detection and Avoidance
		2.6.4. Neighborhood Robotic Fish Perception
	2.7. DISCUSSION
	2.8. CONCLUSION
CHAPTER 3: Boxfish-Like Robot with an Artificial Lateral Line System
	3.1. MULTI-FIN-ACTUATED BOXFISH-LIKE ROBOT
	3.2. CAUDAL-FIN-ACTUATED BOXFISH-LIKE ROBOT
CHAPTER 4: Online State Estimation of a Boxfish-Like Robot Using Artificial Lateral Line System
	4.1. INTRODUCTION
	4.2. MATERIALS AND METHODS
		4.2.1. The Experimental Description
		4.2.2. The Experimental Procedures
	4.3. PRESSURE VARIATION MODEL
		4.3.1. Theoretical Analysis for Hydrodynamic Pressure Variation
		4.3.2. Pressure Variation Models for Multiple Motions of the Robotic Fish
		4.3.3. Identification Process of the Model Parameters
		4.3.4. Pressure Variation Model Based Motion Parameters Estimation
	4.4. ARTIFICIAL LATERAL LINE BASED TRAJECTORY ESTIMATION
		4.4.1. Trajectory Estimation of Rectilinear Motion
		4.4.2. Trajectory Estimation of Turning Motion
		4.4.3. Trajectory Estimation of Gliding Motion
		4.4.4. Trajectory Estimation of Spiral Motion
	4.5. EXPERIMENTS
		4.5.1. Rectilinear Motion
		4.5.2. Turning Motion
		4.5.3. Gliding Motion
		4.5.4. Spiral Motion
		4.5.5. Discussion
	4.6. CONCLUSIONS AND FUTURE WORK
CHAPTER 5: Artificial Lateral Line Based Local Sensing Between Two Adjacent Boxfish-Like Robots
	5.1. INTRODUCTION
	5.2. MATERIALS AND METHODS
		5.2.1. The Experimental Description
			5.2.1.1. Individual Differences Among Pressure Sensors
			5.2.1.2. The Experimental Platform
			5.2.1.3. The Experimental Principle
			5.2.1.4. The Experimental Parameters
			5.2.1.5. The Experimental Procedures
		5.2.2. Computation Fluid Dynamics Simulation
	5.3. EXPERIMENTS
		5.3.1. Experiment 1: Sensing the Relative Vertical Distance Between a Robotic Fish and Its Adjacent Oscillating Caudal Fin
		5.3.2. Experiment 2: Sensing the Oscillating Amplitude of Adjacent Oscillating Caudal Fin
		5.3.3. Experiment 3: Sensing the Oscillating Frequency of Adjacent Oscillating Caudal Fin
		5.3.4. Experiment 4: Sensing the Oscillating Offset of Adjacent Oscillating Caudal Fin
		5.3.5. Experiment 5: Sensing the Relative Yaw Angle Between a Robotic Fish and Its Adjacent Oscillating Caudal Fin
		5.3.6. Experiment 6: Sensing the Relative Pitch Angle Between a Robotic Fish and Its Adjacent Oscillating Caudal Fin
		5.3.7. Experiment 7: Sensing the Relative Roll Angle Between a Robotic Fish and Its Adjacent Oscillating Caudal Fin
	5.4. DISCUSSION
		5.4.1. Further Discussion on the Simplification of Sensing the Relative States Between Two Adjacent Robotic Fish
		5.4.2. The Selection of the Relative Longitudinal Distance Between the Individual Caudal Fin and the Robotic Fish
		5.4.3. Artificial Lateral Line System’s Application in Exploring Reverse Kármán Vortex Street and Kármán Vortex Street
		5.4.4. Artificial Lateral Line System’s Potential in Multiple Underwater Vehicles or Robots Based Underwater Task Execution
	5.5. CONCLUSIONS AND FUTURE WORK
CHAPTER 6: Artificial Lateral Line Based Relative State Estimation for Two Adjacent Boxfish-Like Robots
	6.1. INTRODUCTION
	6.2. EXPERIMENTAL APPROACH
		6.2.1. The Robotic Fish with an Artificial Lateral Line System
		6.2.2. Experimental Description
		6.2.3. Pretreatment of the Data
		6.2.4. Random Forest for Regression Task
		6.2.5. Back Propagation Neural Network (BPNN)
		6.2.6. Support Vector Regression (SVR) and Multivariable Linear Regression (REG)
		6.2.7. Sensitivity of the Pressure Sensors-Measured HPVs to the Relative States
		6.2.8. Importance Measurement of the HPVs Measured by Each Pressure Sensor
		6.2.9. Evaluation of the Regression Model
	6.3. RESULTS
		6.3.1. Insufficiency and Redundancy of the Pressure Sensors
		6.3.2. Regression Results Using the Four Methods
		6.3.3. Random Forest Algorithm Based Relative Yaw Angle Estimation and Oscillating Amplitude Estimation
	6.4. DISCUSSIONS
		6.4.1. Why Have We Focused on Close-Range Sensing?
		6.4.2. The Differences Between Investigating the Relative Yaw Angle Between Two Adjacent Robotic Fish and Investigating the Oscillating Offset of the Upstream Oscillating Caudal Fin
	6.5. CONCLUSIONS AND FUTURE WORK
CHAPTER 7: SUMMARY
Bibliography
Index