Method and system for in situ measurement of root biomass and structure using acoustic waves

Patent metadata

Field	Value
Publication number	US-10352916-B2
Application number	US-201715468005-A
Country	US
Kind code	B2
Filing date	Mar 23, 2017
Priority date	Mar 23, 2017
Publication date	Jul 16, 2019
Grant date	Jul 16, 2019

Abstract

Embodiments of the present invention provide a system and method for accurate, field-ready, non-destructive, and three-dimensional plant root characterization using acoustic signals. The system is portable, fast, precise, and can be used in field conditions, including moist soil, to visualize root structure or mass distribution, without damaging growing crops. The system may also be applied to characterize other underground objects, such as pipes, building foundations, archaeological artifacts, or mineral ores. During operation, the system generates a source acoustic signal. The system sends the source acoustic signal to an actuator acoustically coupled directly to a plant. The system obtains a response acoustic signal from an underground transducer monitoring a root of the plant. The system analyzes the response acoustic signal according to a model and based on the source acoustic signal. The system then determines, based on the analyzed response acoustic signal, a physical configuration of the plant root.

First claim

What is claimed is:

1. A computer-implemented method for measuring plant roots, comprising:
generating, by a computer system, a source acoustic signal;
sending the source acoustic signal to an actuator coupled to a plant, wherein the actuator is configured to generate a local acoustic signal directed to the plant based on the source acoustic signal;
receiving, from a transducer monitoring a root of the plant, a response acoustic signal indicating a response of the local acoustic signal from the root;
analyzing the response acoustic signal based on a model and the source acoustic signal; and
determining, based on the analyzed response acoustic signal, a phenotype of the root.

2. The method of claim 1 , further comprising:
obtaining multiple response acoustic signals from multiple transducers monitoring the root;
analyzing the multiple response acoustic signals based on the model, the source acoustic signal, and a position of a respective transducer, and
determining the phenotype of the root further based on the analyzed multiple response acoustic signals.

3. The method of claim 1 , wherein the model is a machine learning model trained based on a known physical configuration of the root.

4. The method of claim 1 , wherein the model is based on propagation of the source and response acoustic signals underground and within the plant.

5. The method of claim 1 , wherein the phenotype of the root indicates a geometrical structure or a mass distribution of the root.

6. The method of claim 1 , wherein the source acoustic signal comprises one or more of:
a frequency swept standing wave;
a frequency swept pulse;
a chirp;
a signal that is modulated in time domain;
a signal that spans over a continuous range of frequencies; and
a signal that is based on a predetermined number of discrete frequencies.

7. The method of claim 1 , wherein analyzing the response acoustic signal comprises measuring one or more of:
a timing of signal arrival;
a signal strength;
a propagation delay;
a phase shift based on a reference signal;
a frequency shift based on a reference frequency;
a parameter in time domain; and
a parameter in frequency domain.

8. The method of claim 1 , wherein the actuator is configured to generate the local acoustic signal along or perpendicular to a stalk of the plant.

9. An apparatus for measuring plant roots, comprising:
a signal generating module configured to generate a source acoustic signal;
a sending module configured to send the source acoustic signal to an actuator coupled to a plant, wherein the actuator is configured to generate a local acoustic signal directed to the plant based on the source acoustic signal;
an obtaining module configured to receive, from the transducer monitoring the root, a response acoustic signal indicating a response of the local acoustic signal from the root;
an analyzing module configured to analyze the response acoustic signal based on a model and the source acoustic signal; and
a determining module configured to determine, based on the analyzed response acoustic signal, a phenotype of the root.

10. The apparatus of claim 9 , wherein:
the obtaining module is further configured to obtain multiple response acoustic signals from multiple transducers monitoring the root;
the analyzing module is further configured to analyze the multiple response acoustic signals based on the model, the source acoustic signal, and a position of a respective transducer; and
the determining module is further configured to determine the phenotype of the root further based on the analyzed multiple response acoustic signals.

11. The apparatus of claim 9 , wherein the model is a machine learning model trained based on a known physical configuration of the root.

12. The apparatus of claim 9 , wherein the source acoustic signal comprises one or more of:
a frequency swept standing wave;
a frequency swept pulse;
a chirp;
a signal that is modulated in time domain;
a signal that spans over a continuous range of frequencies; and
a signal that is based on a predetermined number of discrete frequencies.

13. The apparatus of claim 9 , wherein while analyzing the response acoustic signal, the analyzing module is further configured to measure one or more of:
a timing of signal arrival;
a signal strength;
a propagation delay;
a phase shift based on a reference signal;
a frequency shift based on a reference frequency;
a parameter in time domain; and
a parameter in frequency domain.

14. A computing system for measuring plant roots, the system comprising:
a set of processors;
a non-transitory computer-readable medium coupled to the set of processors storing instructions thereon that, when executed by the processors, cause the processors to perform a method for measuring plant roots, the method comprising:
generating a source acoustic signal;
sending the source acoustic signal to an actuator coupled to a plant wherein the actuator is configured to generate a local acoustic signal directed to the plant based on the source acoustic signal;
receiving, from the transducer monitoring the root, a response acoustic signal indicating a response of the local acoustic signal from the root;
analyzing the response acoustic signal based on a model and the source acoustic signal; and
determining, based on the analyzed response acoustic signal, a phenotype of the root.

15. The computing system of claim 14 , wherein the method further comprises:
obtaining multiple response acoustic signals from multiple transducers monitoring the root; and
analyzing the multiple response acoustic signals based on the model, the source acoustic signal, and a position of a respective transducer.

16. The computing system of claim 14 , wherein the model is a machine learning model trained based on a known physical configuration of the root.

17. The computing system of claim 14 wherein the phenotype of the root indicates a geometrical structure or a mass distribution of the root.

18. The computing system of claim 14 , wherein the source acoustic signal comprises one or more of:
a frequency swept standing wave;
a frequency swept pulse;
a chirp;
a signal that is modulated in time domain;
a signal that spans over a continuous range of frequencies; and
a signal that is based on a predetermined number of discrete frequencies.

19. The computing system of claim 14 , wherein analyzing the response acoustic signal comprises measuring one or more of:
a timing of signal arrival;
a signal strength;
a propagation delay;
a phase shift based on a reference signal;
a frequency shift based on a reference frequency;
a parameter in time domain; and
a parameter in frequency domain.

20. The computing system of claim 14 , wherein the actuator is configured to generate the local acoustic signal along or perpendicular to a stalk of the plant.

Assignees

• Palo Alto Res Ct Inc

Inventors

• Cocker Eric D
• Elrod Scott A
• Rosa Uriel A
• Rivest Jessica L B
• Daniel George W
• Schwartz David E

Classifications

• G01N29/348

  with frequency characteristics, e.g. single frequency signals, chirp signals (measuring frequency of mechanical vibrations or acoustic waves in general G01H1/06, G01H3/04; measuring frequency or analysing frequency spectra G01R23/00) · CPC title

• G01N2291/011

  Velocity or travel time · CPC title

• G01N2291/101

  one transducer · CPC title

• G01N2291/015

  Attenuation, scattering · CPC title

• G01N29/343

  pulse waves, e.g. particular sequence of pulses, bursts · CPC title