Acoustic Localization and Communication Using a MEMS Microphone for Low-cost and Low-power Bio-inspired Underwater Robots Akshay Hinduja1 Yunsik Ohm1 Jiahe Liao2 Carmel Majidi1 and Michael Kaess2_2

2025-04-30 0 0 8.19MB 8 页 10玖币

侵权投诉

Acoustic Localization and Communication Using a MEMS Microphone

for Low-cost and Low-power Bio-inspired Underwater Robots

Akshay Hinduja1, Yunsik Ohm1, Jiahe Liao2, Carmel Majidi1, and Michael Kaess2

Abstract— Having accurate localization capabilities is one

of the fundamental requirements of autonomous robots. For

underwater vehicles, the choices for effective localization are

limited due to limitations of GPS use in water and poor environ-

mental visibility that makes camera-based methods ineffective.

Popular inertial navigation methods for underwater localization

using Doppler-velocity log sensors, sonar, high-end inertial

navigation systems, or acoustic positioning systems require

bulky expensive hardware which are incompatible with low-

cost, bio-inspired underwater robots. In this paper, we introduce

an approach for underwater robot localization inspired by

GPS methods known as acoustic pseudoranging. Our method

allows us to potentially localize multiple bio-inspired robots

equipped with commonly available micro electro-mechanical

systems microphones. This is achieved through estimating the

time difference of arrival of acoustic signals sent simultaneously

through four speakers with a known constellation geometry.

We also leverage the same acoustic framework to perform one-

way communication with the robot to execute some primitive

motions. To our knowledge, this is the ﬁrst application of the

approach for the on-board localization of small bio-inspired

robots in water. Hardware schematics and the accompanying

code are released to aid further development in the ﬁeld3.

I. INTRODUCTION AND RELATED WORK

Underwater robots are increasingly being used to explore

uncharted depths, inspect artiﬁcial undersea structures, as

well as monitor aquatic life and the physical and chemical

properties of their surrounding environment. While larger

robots like autonomous underwater vehicles (AUVs) are

more suited to tasks such as the inspection of ship hulls,

harbors [1] and sub-sea pipelines, bio-inspired robots are

better equipped to closely monitor underwater life given

their smaller size and lack of moving parts likely to disturb

the surrounding environment [2, 3]. Bio-inspired robots have

improved over recent years in aspects of locomotion, control,

and communication. In 2014, Marchese et al. [4] presented

their novel actuation method for a soft robotic ﬁsh capable

of agile movement. Soon after in 2015, the group presented

a compact acoustic communication module for in-water

control of the robot [5]. More recently, an improvement

over the previous works was presented by Katzschmann et

al. [6] through their soft robotic ﬁsh, SoFi, which could be

controlled using a handheld acoustic modem based controller

presented in the research done by Marchese et al. [5]. The

The authors are with the Department of Mechanical Engineering1,

the Robotics Institute2, Carnegie Mellon University, Pittsburgh, PA

15213, USA. {ahinduja, yohm, jiahel, cmajidi,

kaess}@andrew.cmu.edu

This work was partially supported by the National Oceanographic Part-

nership Program under ONR award N00014-18-1-2843.

https://github.com/rpl-cmu/underwater-acoustic-pseudoranging3

drawback here is that the controller is considerably expensive

and can only communicate with one robot at a time. A more

recent example of a smaller bio-inspired underwater robot is

PATRICK from Patterson et al. [7], which demonstrated the

capability of performing closed-loop locomotion planning

using an external camera and Bluetooth communication.

Here, the use of Bluetooth for passing instructions to the

robot limits it to only operate at surface level or shallow

water since it requires a non-submerged Bluetooth antenna

to receive instructions. However, despite the limitations,

these examples demonstrate the promising transition towards

autonomy for bio-inspired underwater robots.

(a) Setup for pseudorange localization

(b) Bio-inspired ﬁsh robot

Fig. 1: Water Tank Experiments: (a) Experimental veriﬁcation of acoustic

pseudoranging based localization. Estimated positions are compared to

known ground truth positions. (b) A proof-of-concept bio-inspired robot

with three actuated ﬁns connected to the receiver module is veriﬁed to

execute selective motions on receiving speciﬁc acoustic signals.

A fundamental necessity for an autonomous robot to

navigate and perform speciﬁc tasks is the ability to localize

itself accurately. Traditionally in robotics, there are systems

that can achieve accurate location estimates by utilizing

arXiv:2210.01089v1 [cs.RO] 3 Oct 2022

one or more sensors like cameras, lidars, radars, inertial

measurement units (IMUs) and the global positioning system

(GPS). Underwater robots face unique challenges when

it comes to localization. In real-world scenarios, sensors

like cameras and lidar are ineffective due to turbidity, and

signals from absolute positioning systems like GPS cannot

be received underwater. Most commercial AUVs rely on

sensors like Doppler velocity logs (DVLs), depth sensors

and IMUs for inertial navigation methods to help localize

themselves. Sonars and cameras can be used for aiding

inertial navigation by leveraging the geographical features of

the surrounding environment. As mentioned earlier, cameras

are effective only in clear water and sonars often make the

cost of underwater vehicles prohibitively expensive. Costs

and practicality aside, their weights and bulky form factors

limit them to be used only on larger robots. In this work, our

focus is on localization for small, low-cost and low-power,

untethered robots.

For localization of smaller and cheaper underwater robots,

such as bio-inspired robot ﬁsh, acoustic positioning methods

have more potential as a viable solution. The oldest and

most popular methods of acoustic positioning are long base-

line or short baseline (LBL/SBL) and ultra short baseline

(USBL) systems [8, 9]. These methods are two-way travel-

time (TWTT) methods, which means the beacons and the

AUVs need to be equipped with active acoustic systems to

communicate with one another. The use of atomic clocks

to synchronize both beacons and AUVs can allow for

localization using one-way travel-time (OWTT) methods.

Earlier approaches achieved range-only OWTT localization

and navigation by using ﬁltering or fusion with other on-

board sensors [10, 11]. More recently, Rypkema et al. [12]

presented their novel method of one-way travel-time inverted

ultra-short baseline (OWTT-iUSBL), which allowed for real-

time on-board navigation using a single speaker and an

array of four passively listening hydrophones on the AUV.

Matched ﬁltering and phased-array beamforming on the four

received signals from each hydrophone gives a range and

bearing estimate of the robot with respect to the speaker

position.

While the OWTT-iUSBL can perform effective local-

ization with a relatively cost effective setup compared to

commercial AUVs, the components used are still expensive

and too heavy to be used on much smaller bio-inspired

robots. Atomic clocks, which make OWTT methods possible,

are still expensive relative to other options. For a swarm of

robots, these costs can add up signiﬁcantly. Accurate time-

of-arrival information is paramount for OWTT methods, and

cheaper embedded real time clocks (RTCs) are prone to

signiﬁcant drift and lack the superior precision of atomic

clocks. Time-difference of arrival (TDOA) techniques, on the

other hand, allow us to avoid the need to know the exact time

a signal was sent. This is achieved by instead measuring the

differences between the arrival time of multiple signals which

were known to be transmitted simultaneously from different

beacons. This technique is commonly referred to as pseudo-

ranging, and is the foundation of GPS localization [13].

Fig. 2: System Overview: The transmitter module periodically plays a

sequence of identical chirps (black) followed by another, different, chirp

(blue) through a constellation of speakers. The receiver module uses the

information to estimate position as well as execute basic locomotion tasks.

In this paper we present a method for localization of a

small robotic ﬁsh that is based on acoustic pseudoranging.

This is accomplished using cheap, miniaturized, low power

sensors and computation. Referring to Fig. 2, pseudorange

localization and acoustic communication is performed on a

ﬁsh-inspired robot that swims in a water tank that is instru-

mented with acoustic speakers and receivers. The robotic ﬁsh

is embedded with a small and inexpensive micro-electro-

mechanical systems (MEMS) microphone that is used for

communication and localization through the estimation of

TDOA of signals sent simultaneously from the four speakers

in the water tank.

As an acoustically passive method on the receiver side,

pseudoranging allows for multiple agents to be localized

simultaneously without the need for individual signals spec-

ifying time of ﬂight (TOF) information or cross commu-

nication between transmitter and receiver to achieve time

of arrival (TOA). Variations of acoustic pseudoranging have

been utilized for the localization of larger underwater vehi-

cles before. Jorgensen et al. [14] presented an observer to

estimate several parameters like position, velocity and IMU

biases. Leveraging pseudorange measurement differences

along with attitude and accelerometer gave better position

estimates. A long range underwater navigation algorithm

based on acoustic pseudoranging was tested in an area

spanning ~275,000km2by Mikhalevsky et al. [15] by using

GPS assisted beacons. Recent works by Berlinger et al. [16]

and Novak et al. [17] have also explored the localization of

a swarm of ﬁsh, both robotic and natural, through optical

and acoustic methods respectively. However both methods

perform localization from an external observer and not on-

board the agent, which makes acoustic pseudoranging more

suitable to use cases where geo-referenced data collection,

and on-board navigation are important.

II. SYSTEM ARCHITECTURE

This section will now describe the hardware and frame-

work for acoustic pseudoranging based localization and one-

way acoustic communication. A diagrammatic illustration is

shown in Fig. 2. There are a minimum of four speakers

connected to a single computer playing a sequence of chirps

followed by a single, different chirp, periodically. The re-

ceiver is time synchronized with the transmitter computer

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

AcousticLocalizationandCommunicationUsingaMEMSMicrophoneforLow-costandLow-powerBio-inspiredUnderwaterRobotsAkshayHinduja1,YunsikOhm1,JiaheLiao2,CarmelMajidi1,andMichaelKaess2AbstractHavingaccuratelocalizationcapabilitiesisoneofthefundamentalrequirementsofautonomousrobots.Forunderwatervehicles,thech...

展开>> 收起<<

Acoustic Localization and Communication Using a MEMS Microphone for Low-cost and Low-power Bio-inspired Underwater Robots Akshay Hinduja1 Yunsik Ohm1 Jiahe Liao2 Carmel Majidi1 and Michael Kaess2_2.pdf

共8页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Acoustic Localization and Communication Using a MEMS Microphone for Low-cost and Low-power Bio-inspired Underwater Robots Akshay Hinduja1 Yunsik Ohm1 Jiahe Liao2 Carmel Majidi1 and Michael Kaess2_2

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: