ASTM F3265-17(2023)

Standard Test Method for Grid-Video Obstacle Measurement

SIGNIFICANCE AND USE
4.1 Assuming the vehicle stays on its path and an obstacle appears within the stop zone, the vehicle will collide with the obstacle. Even within the stop zone, obstacle detection should cause the vehicle to slow down as early as possible using non-contact sensing or contact bumpers. ANSI/ITSDF B56.5:2012 discusses a test method to detect standard test pieces beyond the minimum vehicle stopping distance at 50 % and 100 % of vehicle rated speeds.  
4.2 This test method can apply to A-UGVs for testing obstacle-sensing capabilities and automatic guided industrial vehicles in automatic mode of operation in non-restricted areas as described in ANSI/ITSDF B56.5.  
4.3 Researchers2, 3 used two-dimensional (2D) laser detection and ranging (LADAR) sensors mounted to an A-UGV. In contrast to the earlier experiments in which the test piece was static, in these experiments the A-UGV and the test piece were both moving. The 2D sensor was mounted to the A-UGV to scan horizontally with the beam approximately 10 cm (4 in.) above and parallel to the floor and confined to detecting the vehicle path (vehicle width) at the maximum stopping distance (coasting or braking). Note that the sensor scan width can be set to any width, including the ANSI/ITSDF B56.5 standard, non-hazard zone vehicle path width of the vehicle plus 0.5 m (1.6 ft). The test piece entered the A-UGV path within the exception zone, was detected by the safety sensor, and the distance of the test piece to the A-UGV and the A-UGV stopping distance measurements were calculated and analyzed.
SCOPE
1.1 This test method measures an automatic/automated/autonomous-unmanned ground vehicle (A-UGV) kinetic energy reduction when objects appear in the A-UGV path and within the stop-detect range of the vehicle safety sensors in situations in which the desired reaction is for the vehicle to stop as opposed to avoiding the obstacle by traveling on an alternative path. The test method measures the performance of the A-UGV only and does not measure the effect on the stability of loads. This test method describes the use of one test piece as described in ANSI/ITSDF B56.5. Other test pieces from ANSI/ITSDF B56.5 could be used. This test method is intended for use by A-UGV manufacturers, installers, and users. This test method does not substitute for required safety testing under ANSI/ITSDF B56.5 or other normative standards.  
1.2 Performing Location—This test method shall be performed in a testing laboratory or the location where the apparatus and environmental test conditions are implemented. Environmental conditions are recorded as specified in Practice F3218.  
1.3 Units—The values stated in SI units are to be regarded as the standard. The values given in parentheses are not precise mathematical conversion to inch-pound units. They are close approximate equivalents for the purpose of specifying material dimensions or quantities that are readily available to avoid excessive fabrication costs of test apparatuses while maintaining repeatability and reproducibility of the test method results. These values given in parentheses are provided for information only and are not considered standard.  
1.4 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use.  
1.5 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

General Information

• Status: Published
• Publication Date: 30-Nov-2023
• ICS: 43.020 - Road vehicles in general
• Technical Committee: F45 - Robotics, Automation, and Autonomous Systems
• Drafting Committee: F45.03 - A-UGV Object Detection and Protection

Relations

• Replaces: ASTM F3265-17 - Standard Test Method for Grid-Video Obstacle Measurement
• Effective Date: 01-Dec-2023
• Referred By: ASTM F3243-21 - Standard Practice for Implementing Communications Impairments on A-UGV Systems
• Effective Date: 01-Dec-2023
• Referred By: ASTM F3470-20 - Standard Guide for A-UGV Capabilities
• Effective Date: 01-Dec-2023

Overview

ASTM F3265-17(2023): Standard Test Method for Grid-Video Obstacle Measurement provides a laboratory-based test procedure for evaluating the obstacle detection and response capabilities of automatic, automated, and autonomous-unmanned ground vehicles (A-UGVs). The standard is developed by ASTM to help manufacturers, installers, and users of A-UGVs measure performance when objects suddenly appear in the A-UGV’s path, within the range of onboard safety sensors.

This test method is critical for enhancing the safe integration of A-UGVs in industrial or semi-restricted environments-especially where collision risks between vehicles and humans need to be minimized.

Key Topics

• Purpose: Measures A-UGV kinetic energy reduction when encountering obstacles within the vehicle’s stop-detect sensor range.
• Applicability: Designed for testing A-UGVs, including automatic guided industrial vehicles used in automatic mode in non-restricted areas.
• Test Approach: Utilizes a grid on the floor, video analysis, and test pieces (typically representing a human leg, as referenced in ANSI/ITSDF B56.5) to assess stopping performance.
• Performance Focus: Determines A-UGV reaction-emphasizing braking and deceleration versus obstacle avoidance.
• Data Capture: Records key distances, timings (e.g., when the test piece enters the stop zone, vehicle stops), and kinetic energy reduction in response to obstacles.
• Statistical Reliability: Specifies required repetitions and statistical confidence (generally at least 90% reliability with 95% confidence for zero-failure tests).

Applications

The primary application of ASTM F3265 is in industrial and commercial settings where robotic vehicles operate independently:

• A-UGV Manufacturer Testing: Supports the development, validation, and demonstration of safety sensing and response algorithms for unmanned vehicles.
• Factory Acceptance and Commissioning: Enables installers and users to verify obstacle response as part of factory acceptance tests, providing documented, repeatable safety validation.
• Regulatory Compliance: Complements safety testing required under ANSI/ITSDF B56.5 and helps demonstrate compliance with safety standards for driverless vehicles.
• Workplace Safety Programs: Assists safety managers in assessing the risk of collisions and refining operational procedures involving A-UGVs.
• Research and Development: Provides a standardized approach for researchers evaluating novel sensor technologies (such as 2D LADAR or other non-contact detection systems) mounted on ground vehicles.

Related Standards

Understanding ASTM F3265 in context is important, as it references and complements other key industry documents:

• ANSI/ITSDF B56.5: Safety Standard for Driverless, Automatic Guided Industrial Vehicles and Automated Functions of Manned Industrial Vehicles. Offers broader requirements for safe vehicle operation and stopping distances.
• ASTM F3200: Terminology for Robotics, Automation, and Autonomous Systems. Establishes commonly used terms for clarity in communication.
• ASTM F3218: Practice for Documenting Environmental Conditions for Utilization with A-UGV Test Methods. Guides the recording of test environments to ensure reproducibility and comparability.

Practical Value

Adopting ASTM F3265-17(2023) delivers significant value across the robotics and automation industry:

• Consistency: Establishes a uniform, repeatable, and reliable process to benchmark A-UGV obstacle sensing and stopping performance.
• Safety Assurance: Enhances protection of personnel working near A-UGVs by verifying the effectiveness of obstacle response systems.
• Risk Reduction: Identifies potential performance gaps or hazards before deployment, reducing workplace incidents and equipment damage.
• Market Confidence: Demonstrates due diligence to customers, regulators, and insurers through adherence to internationally recognized procedures.

Keywords: A-UGV, unmanned ground vehicle, obstacle detection, grid-video measurement, safety stop, industrial robotics, kinetic energy reduction, automated guided vehicle, ASTM F3265, ANSI/ITSDF B56.5, safety standards.

Standards Content (Sample)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: F3265 − 17 (Reapproved 2023)
Standard Test Method for
Grid-Video Obstacle Measurement
This standard is issued under the fixed designation F3265; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
INTRODUCTION
Safe control of automatic/automated/autonomous-unmanned ground vehicles (A-UGVs) is critical
in industrial environments where workers are or may be present. A-UGV safe control is typically
based on sensors that detect stationary standard test pieces (used in ANSI/ITSDF B56.5) representing
2, 3
humans. This and other test method developments have been experimented and published. The
experimental results were used to recommend improvements to the ANSI/ITSDF B56.5 safety
standard stopping distance exception language in 2014. Subcommittee consensus changed ANSI/
ITSDF B56.5 to make it mandatory to reduce vehicle kinetic energy should an object (for example,
person, materials, or equipment) appear in the vehicle path and within the stop detect range of the
vehicle safety sensors. The language that has been proposed as an amendment to the ANSI/ITSDF
B56.5 standard is: “Should an object suddenly appear in the path of the vehicle between the leading
edge of the sensing field and the vehicle (for example, an object falling from overhead or a pedestrian
stepping into the path of a vehicle at the last instant), the vehicle shall initiate braking in accordance
with brake system (see 8.8.1), but may not be expected to stop in time to prevent contact with object.”
While manufacturers of A-UGVs may have access to internal system logs and data that demonstrate
the successful initiation of braking as required, users may not have access to that information. This test
method provides an optional, standard performance test method for A-UGVs to enable industrial
vehicle manufacturers and users to implement a common test to demonstrate expected vehicle
operation in the case of objects appearing in the A-UGV path and within the stop-detect range of the
vehicle safety sensors.
1. Scope stability of loads. This test method describes the use of one test
piece as described in ANSI/ITSDF B56.5. Other test pieces
1.1 This test method measures an automatic/automated/
from ANSI/ITSDF B56.5 could be used. This test method is
autonomous-unmanned ground vehicle (A-UGV) kinetic en-
intended for use by A-UGV manufacturers, installers, and
ergy reduction when objects appear in the A-UGV path and
users. This test method does not substitute for required safety
within the stop-detect range of the vehicle safety sensors in
testing under ANSI/ITSDF B56.5 or other normative standards.
situations in which the desired reaction is for the vehicle to stop
as opposed to avoiding the obstacle by traveling on an
1.2 Performing Location—This test method shall be per-
alternative path. The test method measures the performance of
formed in a testing laboratory or the location where the
the A-UGV only and does not measure the effect on the
apparatus and environmental test conditions are implemented.
Environmental conditions are recorded as specified in Practice
This test method is under the jurisdiction of ASTM Committee F45 on
F3218.
Robotics, Automation, and Autonomous Systems and is the direct responsibility of
Subcommittee F45.03 on A-UGV Object Detection and Protection.
1.3 Units—The values stated in SI units are to be regarded
Current edition approved Dec. 1, 2023. Published January 2024. Originally
as the standard. The values given in parentheses are not precise
approved in 2017. Last previous edition approved in 2017 as F3265 – 17. DOI:
10.1520/F3265-17R23. mathematical conversion to inch-pound units. They are close
Bostelman, Roger, Shackleford, Will, Cheok, Geraldine, and Saidi, Kamel,
approximate equivalents for the purpose of specifying material
“Safe Control of Manufacturing Vehicles Research Towards Standard Test
dimensions or quantities that are readily available to avoid
Methods,” Progress in Material Handling Practice, Book Chapter, June 2012.
Bostelman, Roger, Norcross, Richard, Falco, Joe, and Marvel, Jeremy, “Devel- excessive fabrication costs of test apparatuses while maintain-
opment of Standard Test Methods for Unmanned and Manned Industrial Vehicles
ing repeatability and reproducibility of the test method results.
Used Near Humans,” SPIE 2013, Baltimore, Maryland, May 2013.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
F3265 − 17 (2023)
These values given in parentheses are provided for information 3.2.8 stop zone, n—the area in front of the direction of travel
only and are not considered standard. of the A-UGV where activation of the obstruction sensor
causes a safety stop of the vehicle as per ANSI/ITSDF B56.5.
1.4 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the 3.2.9 stopping distance, n—distance required for vehicle to
responsibility of the user of this standard to establish appro- stop after detecting obstruction.
priate safety, health, and environmental practices and deter-
3.2.10 repetition, n—performance of a task.
mine the applicability of regulatory limitations prior to use.
3.2.11 task, n—sequence of movements and measurements
1.5 This international standard was developed in accor-
that compromise one repetition within a test.
dance with internationally recognized principles on standard-
3.2.12 test, n—a collection of task repetitions.
ization established in the Decision on Principles for the
Development of International Standards, Guides and Recom-
4. Significance and Use
mendations issued by the World Trade Organization Technical
4.1 Assuming the vehicle stays on its path and an obstacle
Barriers to Trade (TBT) Committee.
appears within the stop zone, the vehicle will collide with the
obstacle. Even within the stop zone, obstacle detection should
2. Referenced Documents
cause the vehicle to slow down as early as possible using
2.1 ASTM Standards:
non-contact sensing or contact bumpers. ANSI/ITSDF
F3200 Terminology for Robotics, Automation, and Autono-
B56.5:2012 discusses a test method to detect standard test
mous Systems
pieces beyond the minimum vehicle stopping distance at 50 %
F3218 Practice for Documenting Environmental Conditions
and 100 % of vehicle rated speeds.
for Utilization with A-UGV Test Methods
4.2 This test method can apply to A-UGVs for testing
2.2 ANSI/ITSDF Standard:
obstacle-sensing capabilities and automatic guided industrial
ANSI/ITSDF B56.5 Safety Standard for Driverless, Auto-
vehicles in automatic mode of operation in non-restricted areas
matic Guided Industrial Vehicles and Automated Func-
as described in ANSI/ITSDF B56.5.
tions of Manned Industrial Vehicles
2, 3
4.3 Researchers used two-dimensional (2D) laser detec-
tion and ranging (LADAR) sensors mounted to an A-UGV. In
3. Terminology
contrast to the earlier experiments in which the test piece was
3.1 Terms not defined herein are defined in Terminology
static, in these experiments the A-UGV and the test piece were
F3200.
both moving. The 2D sensor was mounted to the A-UGV to
3.2 Definitions of Terms Specific to This Standard:
scan horizontally with the beam approximately 10 cm (4 in.)
3.2.1 collide time, n—when the automatic/automated/
above and parallel to the floor and confined to detecting the
autonomous-unmanned ground vehicle (A-UGV) collides with
vehicle path (vehicle width) at the maximum stopping distance
the test piece.
(coasting or braking). Note that the sensor scan width can be
3.2.2 defined areas, n—space constrained by test method set to any width, including the ANSI/ITSDF B56.5 standard,
boundaries for A-unmanned ground vehicle (A-UGV) opera-
non-hazard zone vehicle path width of the vehicle plus 0.5 m
tion. (1.6 ft). The test piece entered the A-UGV path within the
exception zone, was detected by the safety sensor, and the
3.2.3 enter time, n—when the test piece enters the stop zone
distance of the test piece to the A-UGV and the A-UGV
triggering photosensor 1.
stopping distance measurements were calculated and analyzed.
3.2.4 start line, n—line across the path of the vehicle used to
signal when the test piece can be inserted into the stop 5. Apparatus
detection range as measured in 7.3.
5.1 List of Materials:
3.2.5 start location, n—the initial zero velocity position of
5.1.1 Grid printed on paper or marked on the floor at regular
A-UGV at beginning of each test repetition; the start location
intervals.
should be at a point from which the vehicle can accelerate up
5.1.2 Photosensors 1 and 2.
to test speed before the leading edge crosses the start line along
5.1.3 Lights 1 and 2.
the test trajectory.
5.1.4 Video camera and recorder with nominal 30 frames
per second (fps) frame rate or higher.
3.2.6 start time, n—when the A-UGV crosses the start line
5.1.5 Straight bar and clamp (or optional length of rope or
while traveling at speed.
string, or both, not shown in Fig. 1).
3.2.7 stop time, n—when the A-UGV stops because of
5.1.6 A-UGV.
detection of the test piece.
5.1.7 Vertical cylinder test piece, 70 mm (2.75 in.) diameter
by 400 mm (16 in.), as described in safety standards and
representing a human leg.
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
5.1.8 Onboard A-UGV camera or sighting scope (optional)
contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on (see Fig. 2).
the ASTM website.
5.1.9 Timer to be placed in camera field of view.
Available from the Industrial Truck Standards Development Foundation
(ITSDF), 1750 K St. Nw, Suite 460, Washington, DC 20006, www.itsdf.org. 5.2 Experimental Setup:
F3265 − 17 (2023)
FIG. 1 Apparatus for Grid-Video Test Method
FIG. 2 A-UGV Start Location Calibration Setup with Aligned Posts Using the Onboard Camera FOV and Clamped Bar on the A-UGV
Rear Bumper Pointing Down to a Start Location Marked on the Floor
5.2.1 A printed, grid on paper, taped to the floor, or other 5.2.2 A 70 mm (2.75 in.) diameter by 400 mm (16 in.) tall
measurement marks, shall be next to and as close to the vertical cylinder test piece as specified in safety standards shall
A-UGV as possible along its path without changing vehicle be mounted on a short cart with wheels or other low friction
performance. The grid shall be at least 4 m (13 ft) long by means so that it can be easily pushed into the A-UGV path
0.25 m (0.8 ft) wide and divided into 5 cm (2 in.) segments. without tipping. Optionally, the test piece may be attached to a
The grid should also be labeled every 1 m (3 ft) to provide rope suspended from overhead. The rope should be attached to
additional location information. one end of the test piece so that the piece hangs vertically and
F3265 − 17 (2023)
hangs just short of making contact with the floor. Setting the 6. Hazards
rope vertically, with the target in position, a length of rope is
6.1 In addition to the requirements of 1.4, which addresses
attached to the top of the target and extended horizontally to
the human safety and health concerns, users of this test method
the tester location.
shall also address the equipment preservation concerns and
5.2.3 A video camera shall be mounted in a fixed location,
human A-UGV coexistence concerns.
with the image plane parallel to the travel surface and aligned
NOTE 1—The test requestor and test supervisor agree upon and have the
with the A-UGV path to capture simultaneously both the test
authority to decide upon the environmental conditions under which the
piece and A-UGV motions and stop positions. The video
test(s) is/are to be conducted. Such conditions can be stressful not only to
camera shall have a high enough resolution to capture simul-
the humans but also to the A-UGVs, such as, high or low temperatures,
taneously and clearly, frame-by-frame, the floor grid, test
excessive moisture, and rough terrain that can damage the A-UGV
components or cause unexpected A-UGV motions. Testing of an A-UGV
piece, and A-UGV throughout the test piece motions and stops
may result in exposing the A-UGV, the test area and equipment, and
and the location where the A-UGV detects test piece motion
observers to extraordinary risks. In addition to any other warnings or
and subsequently travels and stops or decelerates. Video
concerns, the test designer shall include a safety plan specific to the
recording shall continue for at least 5 s after the full A-UGV
A-UGV being tested and the test method being used. This plan shall be
crosses the point of entry of the test piece or the A-UGV stops. briefed to all personnel involved and shall include an emergency response
plan should an uncontrolled event occur.
5.2.4 Along the path, a photosensor, Photosensor 1, shall be
placed on or within 400 mm (16 in.) above the floor next to the
7. Calibration
A-UGV so that the emitted beam is along the edge of the
A-UGV stop zone and detects the crossing test piece. The 7.1 Calibration of the A-UGV and Start Location:
emitted beam shall reflect back to the photosensor by a 7.1.1 The A-UGV shall be calibrated to move at 50 % and
reflector placed beyond the A-UGV stop zone. Photosensor 1 100 % of rated speed before testing.
shall control a light, Light 1, on/off that is pointed towards the 7.1.2 The A-UGV shall be positioned at the same start
video camera upon detection of the test piece crossing into the location and orientation for each trial. A calibration method,
A-UGV path. Light 1 shall be simultaneously detected by the shown in Fig. 2, independent of A-UGV sensors, shall be used
video camera during the test. This simplifies identifying the to determine vehicle pose. An example method includes the use
time that the test piece crosses into the stop zone boundary. of a video camera or sighting scope mounted to the vehicle and
5.2.5 Similarly, the beam from a second photosensor, Pho- two posts spaced at approximately 5 m and 10 m along a line
tosensor 2, shall cross the A-UGV path to detect the approach- at an angle other than 0° and 90° from the A-UGV. When the
ing A-UGV and shall be used to turn on a second light, Light A-UGV is at the start location, the two posts are aligned in the
2. Light 2 shall be simultaneously detected by the video camera camera’s field of view (FOV). Also, a thin bar is clamped to the
during the test. Again, this simplifies identifying the time the rear A-UGV bumper pointing down to a spot marked on the
vehicle enters the test area. floor. This ensures that the vehicle was positioned at the same
5.2.6 A test technician and an A-UGV operator are normally start location while the camera/post setup was used to ensure
required to implement the test method. proper vehicle orientation.
5.2.7 If the A-UGV ground clearance is greater than 70 mm, 7.1.3 Tape lines on the floor or other start location calibra-
a front shroud should be added to the A-UGV so that the test tion means can also be used, so long as they allow repeatable
piece cannot roll beneath the vehicle. setup of start locations and orientation of the vehicle. At least
5.2.8 Examples of the test setup are shown in Figs. 1-4. two points of alignment are required to ensure the vehicle is
FIG. 3 Overhead View of the A-UGV Test Piece Positions, Distance Variables, and Moments in Time to be Recorded
F3265 − 17 (2023)
FIG. 4 Snapshots from the Overhead Video Showing when: (a) the Test Piece Crosses the Photosensor, (b) the Test Piece is Detected
by the A-UGV Safety Sensor, (c) the A-UGV Hits the Test Piece (or when the Test Piece Stops in the A-UGV Path), and (d) the A-UGV
Stops
positioned and oriented properly. The alignment points should 0.01 s intervals such as an application on a cellphone, personal
be separated as far apart as possible to minimize Abbe errors computer (PC) display, or a timer.
in vehicle orientation. Also, the alignment points on the vehicle
7.3 Measurement of Stop Field Length at Test Speeds of
and the markings on the floor should be small to minimize
50 % and 100 % of Rated Speed:
parallax effects during positioning.
7.3.1 A static test piece shall be placed in the A-UGV path
7.2 Calibration of the Camera with Grid—Because the grid
and the A-UGV stopping distance checked to ensure that the
is visible in all of the images, calibration of the camera
A-UGV stops before contacting the test piece. The pre-test
distortion is not required. To determine the speed of the
shall be performed five times at A-UGV speeds of 50 % of
vehicle, however, checking the frame rate of the cameras is
rated A-UGV speed and 100 % of rated A-UGV speed. This
required. The camera shall be set to run at a specified frame
pre-test is also used to determine the test piece detection or stop
rate of at least 29 fps. A timer display shall be placed in the
distance width, or both, along the A-UGV path within which
field of view of the camera and set to run. Images shall be
the test piece is inserted as described in the procedure in
captured for at least 10 s. The collected images shall be
Section 8.
reviewed, and the averages time intervals among all images
will be calculated to be within 5 % of the selected rate, with no
8. Procedure
single-second long period more than 10 % off. For example, in
8.1 Pre-Test Information Collection—For data traceability
10 s a camera set at 30 fps collects 310 images for an average
and organization purposes, the test supervisor shall obtain and
of 31 frames per second. This is within 1/30 = 0.033, or 3.3 %.
record the pre-test information first using the form shown in
The number of frames for any single second’s worth of images
Section 10. 8.1.1 – 8.1.14 will assist the test supervisor in
shall not exceed or be below 10 % of the expected rate. In this
completing this form.
case, it would need to be between 27 fps and 33 fps. This
method allo
...