Skip to main content

Python wrapper for Ability Hand API

Project description

Python Ability Hand Wrapper

This PSYONIC Ability Hand wrapper is a asynchronous multithreaded library that can send position, velocity, current and duty messages to the hand and parse its feedback.

Install Pip Package

Preferably using a virtual env install the pip package using:

python3 -m pip install ability-hand

Enable UART & Byte Stuffing using App

The Python wrapper uses serial communication and requires byte stuffing to be enabled. It is recommended to use byte stuffing, fore more info see section 3.5.5 in the API Documentation.

To enable UART and byte stuffing from the PSYONIC app select:

Scan ➡️ SELECT HAND ➡️ Gear Icon ⚙️ (Top Right) ➡️ Troubleshoot ➡️ Developer Mode

and issue the following commands individually.

We16
We46
We47

Run examples

Motor Example With Postion Velcocity and Torque Feedback

Run the motor plot example by entering the following into a terminal:

Linux / macOS

plot_motors

Windows (If the above does not work)

python3 -m ah_examples.plot_motors

Touch Sensors Example

The Ability Hand also has 30 touch sensors (6 per finger) which you can see responding in real time by running the example

Linux / macOS

plot_touch_sensors

Windows (If the above does not work)

python3 -m ah_examples.plot_touch_sensors

Below is a map of each sensor and it's associated plot color.

isolated

Motor Example Without Plots

Running examples with plotting in real time using python uses alot of processing power, and you may see some stutters. Run the hand wave without plots using.

Linux / macOS

hand_wave

Windows (If the above does not work)

python3 -m ah_examples.hand_wave

You can press Ctrl + c to stop

Run in Terminal / Integrate with Code

You can look at the AHSerialClient documentation to see available commands and arguments associated with the class but below is a break-down and examples you can run.


Create Client

First open a python terminal and import the wrapper and AHSerialClient class

from ah_wrapper import AHSerialClient

and create a client instance:

client = AHSerialClient()

This will automatically find a serial connection to the hand and start a thread for writing and reading data from the hand. Any time you issue a set function (i.e. set_position, set_duty, set_command) the command that the write thread issues will be updated, i.e. the client will repeatedly send the most recent target / command. The hand must receive a command every 300ms or else it will exit its API control mode.


Set and Send Position Targets

Update the command with a position target using:

client.set_position([50,50,50,50,50,-50])

This will set all fingers to this desired position. Note that finger indexes always follow the following pattern [index, middle, ring, pinky, thumb flexor, thumb rotator]. Note that the thumb rotator operates in a [0,-100] range and the rest of the fingers are in a [0,100] range.


Understanding Hand class

You can ensure that the feedback from the hand class is correct by issuing:

client.hand.get_position()

Each client has its own Hand class. You can see all the public Hand class commands for reading values here


Set and Send Duty Targets

You can also pass a single argument to a set command, and it will apply it to each of the fingers. For example to set a -10% duty (valid range is [-100,100]) to all fingers, issue.

client.set_duty(-10)

Any of the set commands can take either one argument or will take an array of six values as seen above. If you pass a single value it will apply it to all fingers equally.


Set and Send Torque Targets

Try sending a small amount of torque and notice that the fingers will stop moving if anything gets in their way.

client.set_torque([0.1,0.1,0.1,0.1,0.1,-0.1])

client.set_torque(-0.1)

You can also verify the current draw is not going over the desired set_torque amount.

client.hand.get_current()


Reply Modes

If you issue a client.hand.get_velocity() command you will notice it returns nothing, this is because we have only been sending reply_mode=0 as default, and we have never received any velocity feedback.

The reply modes are as following starting with a 0 index.

  1. Finger position, current, touch sensors
  2. Finger position, rotor velocity, touch sensor
  3. Finger position, current, rotor velocity

Simply pass the reply_mode argument to any set_ command.

client.set_position(30, reply_mode=1)


Send Grip Commands

Lastly you can send individual grip commands from the datasheet.

client.set_grip(0x10)


Closing Client

Lastly finish any session using

client.close()

This will stop any threads sending and receiving commands to the ability hand.


Further examples

In most typical integrations you will create your own write thread and send messages manually using AHSerialClient.send_command() see hand_wave.py or plot_motors.py for examples on how to do that.

Note that in this above example a write thread was constantly sending the current target at 500hz and a read thread was parsing feedback responses. See AHSerialClient for more details

You can update the target that send_command sends using any of the set_ commands, or you can build the message manually and pass that as an argument to the send command.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ability_hand-0.1.3.tar.gz (53.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

ability_hand-0.1.3-py3-none-any.whl (57.2 kB view details)

Uploaded Python 3

File details

Details for the file ability_hand-0.1.3.tar.gz.

File metadata

  • Download URL: ability_hand-0.1.3.tar.gz
  • Upload date:
  • Size: 53.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.10.12

File hashes

Hashes for ability_hand-0.1.3.tar.gz
Algorithm Hash digest
SHA256 78f2bcd09aa8ca1673e4c0350e36bffc911d279201f054ace8ba12babdd5672b
MD5 a28853fff14c2f5576be001b8b486de7
BLAKE2b-256 257aa0bfe86493b8b093dde33c09df004c48d541b6957989b08632cd86b9c03c

See more details on using hashes here.

File details

Details for the file ability_hand-0.1.3-py3-none-any.whl.

File metadata

  • Download URL: ability_hand-0.1.3-py3-none-any.whl
  • Upload date:
  • Size: 57.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.10.12

File hashes

Hashes for ability_hand-0.1.3-py3-none-any.whl
Algorithm Hash digest
SHA256 a025c8117a6a9feac0c012ef1331be0d689cbd296b767ada84caf080b8107c3b
MD5 fc7464148ff4382391aceb41c4dd5651
BLAKE2b-256 eabae4650dcab13c540e97454079bbc3826a05b660cada7fd796560f0b3efe3d

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page