One of the most exciting parts of being in the robotics field is the prospect of what’s to come: the frontiers not yet explored, the stones not yet turned, the breakthroughs not yet discovered. In a world where so much seems “old hat,” it’s refreshing to be a part of the “what’s to come” in technology.
That’s why I’m particularly excited about our latest improvements to the Sawyer software development kit (SDK). This software release pushes the envelope as to what’s available in a collaborative research robot while increasing their accessibility and opening up additional opportunities for universities and corporate research labs. Let’s dive into what the new Sawyer SDK includes:
Integration with Gazebo Physics Simulator
If you’re familiar with Gazebo, you know it’s one of the best robotics simulators in the open source software world. Gazebo integrates well with ROS, the open software framework that Sawyer SDK was built on, making it ideal for running algorithms on a research station before you test it on a physical robot. Ultimately, this lets researchers stretch their Sawyer robot across all lab members: Multiple people can write code and test it in simulation before needing to plug into Sawyer. As always, the Gazebo/Sawyer interface is completely open source, because we believe in providing maximum access to students.
New Motion Interface
Gazebo integration isn’t the only new Sawyer SDK feature. Now you can program the robot at a higher level of motion command: in both Joint and Cartesian space. No longer are you required to program an entire joint trajectory; instead, you can dictate waypoint positions for the end point, and the robot will figure out the required joint angles to achieve each pose. This development significantly lowers the barrier for many researchers, especially those without a full robotics or programming background.
Manufacturing Sawyer: Meet SDK Sawyer
In this release, we brought a few of our manufacturing robot features to the research robot. In particular, we have created interfaces to Intera’s Interaction Controller. This controller runs onboard the robot and allows researchers to program Sawyer using two control modes: Impedance and Force Control. Force and impedance control are useful when robots are about to interact with the environment and manipulate objects. Force control enables the robot to push against objects in the environment with various levels of force or torque. This can be used for different applications, like inserting an object into a fixture.
Impedance control allows the user to control the level of compliance of a robot. This is useful when robots interact with objects in environments that are faced with variabilities. In such a situation, the compliance provided by impedance control allows the robot to adapt to the variability and complete the task in the face of unknown positions. This is analogous to how humans interact with objects. For example, when we are about to insert a key into a lock, we often do not precisely place the key in the perfect position to insert into the lock. Rather, we place the key in the vicinity of the lock, and by making our hand compliant, we allow the lock to guide the key into the correct configuration for insertion. Similarly robots can use impedance control along with fixtures in the environment to achieve their tasks.
We’ve also integrated our ClickSmart family of gripper kits, which allows researchers to pop on a variety of end effectors and have them actuate automatically through the SDK.
It’s impossible to predict what new developments will emerge as a result of these Sawyer SDK upgrades – that’s what keeps roboticists like me enthusiastic about the research space. Regardless of what innovations Sawyer SDK helps unlock, Rethink Robotics will continue to develop new features that empower our research community to keep pushing boundaries and searching for that next big breakthrough.
The post sawyer sdk: empowering research pioneers to blaze new trails appeared first on Rethink Robotics.