# US 2021/0232489 A1 · Method for validating a software

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0232489 A1
- Document ID: US20210232489A1
- Status: Published application
- Kind: A1
- Application number: 17/147,141
- Filed: 2021-01-12
- Published: 2021-07-29
- Earliest claimed: 2020-01-23
- Assignees: Robert Bosch GmbH
- Applicants: Robert Bosch GmbH
- Inventors: Peter Bakucz; Balazs Leidecker; Davut Cimsit
- Web page: https://patentgrep.com/patent/US20210232489A1

## CPC

- G05B 13/0265: {the criterion being a learning criterion}
- G06F 11/3616: {using software metrics}
- G05D 1/0257: {using a radar (radar systems designed for anti-collision purposes between land vehicles or between land vehicle and fixed obstacles G01S13/931)}
- G05D 1/247: using signals provided by artificial sources external to the vehicle, e.g. navigation beacons
- G06F 11/3612: {by runtime analysis (performance monitoring G06F11/3466)}
- G06F 11/3698: Environments for analysis, debugging or testing of software
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- H03H 21/0012: {Digital adaptive filters}
- H04L 5/0048: {Allocation of pilot signals, i.e. of signals known to the receiver (allocation of control signalling H04L5/0053; use of control signalling H04L5/0091)}
- B60W 2050/0052: {Filtering, filters}
- G06F 17/17: Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method ({G06F17/18 takes precedence } ; interpolation for numerical control G05B19/18)
- H03H 17/06: Non-recursive filters
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G05D: SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H03H: IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS (waveguides, resonators, lines or other devices of the waveguide type H01P)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT

## Abstract

A method for validating a software, particularly a driver-assistance software. The method includes receiving of a sensor signal that is to be processed by the software; determining of a reference signal by an adaptive filter based on the sensor signal, the reference signal representing an anticipated sensor signal; determining an error signal based on the sensor signal and the reference signal, the error signal representing a performance of the software; determining an anti-sensor signal by a machine-learning unit based on the sensor signal, the machine-learning unit being trained with sensor signals already evaluated; controlling of the adaptive filter by a control unit based on the determined error signal and the adaptive anti-sensor signal; and validation of the software based on the determined error signal.

## Claims

### Claim 1

A method for validating a software, comprising the following steps:
receiving a sensor signal that is to be processed by the software;
determining, by an adaptive filter, a reference signal based on the sensor signal, the reference signal representing an anticipated sensor signal;
determining an error signal based on the sensor signal and the reference signal, the error signal representing a performance of the software;
determining, by a machine-learning unit, an anti-sensor signal based on the sensor signal, the machine-learning unit being trained with sensor signals already evaluated;
controlling, by a control unit, the adaptive filter based on the determined error signal and the determined anti-sensor signal; and
validating the software based on the determined error signal.

### Claim 2 (depends on 1)

The method as recited in claim 1, wherein the software is a driver-assistance software.

### Claim 3 (depends on 1)

The method as recited in claim 1, wherein the adaptive filter is actively controlled by the control unit.

### Claim 4 (depends on 1)

The method as recited in claim 1, wherein the control unit uses a modified least-mean-square (LMS) algorithm as a control algorithm for controlling the adaptive filter.

### Claim 5 (depends on 1)

The method as recited in claim 1, wherein the performance of the software is determined by a previously specified number of iteration steps of the controlling of the adaptive filter.

### Claim 6 (depends on 1)

The method as recited in claim 1, further comprising the following steps:
detecting of an outlier in the sensor signal by the machine-learning unit; and
rejecting a corresponding anti-sensor signal and a corresponding error signal by the control unit.

### Claim 7 (depends on 1)

The method as recited in claim 1, further comprising the following steps:
executing the method in parallel for at least a first sensor signal and a second sensor signal of different sensors;
detecting an error signal above a previously determined limit value in the case of the first sensor signal, an error signal of the second sensor signal lying below the previously determined limit value; and
rejecting the error signal of the first sensor signal by the control unit.

### Claim 8

A device configured to validate a software, the device configured to:
receive a sensor signal that is to be processed by the software;
determine, by an adaptive filter, a reference signal based on the sensor signal, the reference signal representing an anticipated sensor signal;
determine an error signal based on the sensor signal and the reference signal, the error signal representing a performance of the software;
determine, by a machine-learning unit, an anti-sensor signal based on the sensor signal, the machine-learning unit being trained with sensor signals already evaluated;
control, by a control unit, the adaptive filter based on the determined error signal and the determined anti-sensor signal; and
validate the software based on the determined error signal.

### Claim 9

A control method for an at least semi-autonomous robot, comprising the following steps:
receiving sensor data of the at least semi-autonomous robot for a driver-assistance system;
adapting the driver-assistance system based on a software performance of the driver-assistance system, the software performance being determined by:
receiving a sensor signal that is to be processed by the software,
determining, by an adaptive filter, a reference signal based on the sensor signal, the reference signal representing an anticipated sensor signal,
determining an error signal based on the sensor signal and the reference signal, the error signal representing a performance of the software,
determining, by a machine-learning unit, an anti-sensor signal based on the sensor signal, the machine-learning unit being trained with sensor signals already evaluated,
controlling, by a control unit, the adaptive filter based on the determined error signal and the determined anti-sensor signal, and
validating the software based on the determined error signal; and
controlling of the at least semi-autonomous robot as a function of the adapted driver-assistance system.

### Claim 10

A non-transitory machine-readable storage medium on which is stored a computer program for validating a software, the computer program, when executed by a computer, causing the computer to perform the following steps:
receiving a sensor signal that is to be processed by the software;
determining, by an adaptive filter, a reference signal based on the sensor signal, the reference signal representing an anticipated sensor signal;
determining an error signal based on the sensor signal and the reference signal, the error signal representing a performance of the software;
determining, by a machine-learning unit, an anti-sensor signal based on the sensor signal, the machine-learning unit being trained with sensor signals already evaluated;
controlling, by a control unit, the adaptive filter based on the determined error signal and the determined anti-sensor signal; and
validating the software based on the determined error signal.
