Our mission is to replace mining with reuse as the largest source of the next billion batteries. The problem is that batteries are expensive to reuse today because of non-standardized testing, expensive shipping, and dangerous disassembly by hand. We collect and sort old batteries, test how good they still are, and take them apart. We do this for all types of EV batteries using ML-driven analytics & intelligent robotics. Then we turn them into modular, directly accessible, and low-cost grid storage systems. If we can’t do this, we prepare them for recycling to turn them into raw materials to make new cells. Our vision is to build local factories around the world to become the largest manufacturer of used batteries and their raw materials to realize a truly clean energy revolution.
The Bonobo Robot
To achieve this mission, we build cognitive robots called Bonobo that automatically diagnose, discharge and disassemble EV batteries using ML-based analytics and intelligent robotics. The first-generation robotic system will automatically assess the battery’s state of health, perform safe discharging, and remove covers from arbitrary battery packs (500kg). It then disassembles these batteries from the pack level (500 kg) down to the module- level (25 kg). Bonobo can take batteries apart 4x faster and safer than a human at 6x the throughput, leading to 50% lower unit economics.
Job objective
You will architect, engineer, and deploy the Bonobo software that decides and plans all actions required to perform battery handling for discharging and disassembly of incoming battery packs for robots capturing rich sensor data. You also support designing mechatronic systems integrating industrial robotic systems and gantry-based motion systems to enable the automated battery disassembly process. To do this, you will simulate all ranges of robotic actions to enable the battery disassembly process including mission planning of the disassembly sequence, control algorithms, path planning, and collision avoidance. It does not stop with the robotic system itself, you will craft ambidextrous end-effectors to provide all ranges of disassembly actions. You will validate these functions in simulation and the real world with force controls and feedback.
How you will contribute
- Deliver the Bonobo robotic software system by integrating path planning, control software, and image recognition
- Develop a control software to deploy all functionalities in the robotic system from path planning, positioning, and controlling end-effectors with inverse kinematics.
- Develop mechatronic design for ambidextrous end-effectors and their motion planning and control algorithms. Integrate force control sensors and implement sensor fusion methods.
- Make a selection through trade studies of industrial robotic systems (like Fanuc, Kuka, ABB, Comau), design gantry-based motion systems, and industrial PLC controllers.
- Contribute to the robustness and stability of the robotic system by verifying and validating pipelines through hardware-in-the-loop testing.
- Integrate, prototype, and test simulated and real industrial robot systems.
- Export control routines from robotics simulation software to embedded devices.
The skills & experience that you bring
- At least a B.Sc. in Electronical-, Mechanical-, Controls Engineering, Applied Mathematics, or a similar field.
- Academic background in control theory, sensor fusion, digital signal processing, robotic motion planning, inverse kinematics, collision detection, and avoidance.
- Expert in robotic path planning, trajectory smoothing, and feedback controls within a real product or manufacturing environment.
- 3+ yrs experience in designing control algorithms applied to Classical, Multivariable Feedback, Model Predictive or Nonlinear controls.
- 3+ yrs experience in developing software with strong skills in C/C++, Python, and Matlab-Simulink and have developed software from architecture to production-level code in robotic environments.
- 3+ yrs experience in robotic software development and simulation in ROS, ROS-I, MoveIt for path planning and inverse Kinematics, Gazebo, and/or RobotStudio for simulation.
- 1-3 yrs Experience with OpenCV, Open3D, and PCL libraries.
- Hands-on product experience with industrial articulated robots like Kuka, Fanuc, ABB, and Staubli while performing calibration and motion control on these systems.
- Developed and integrated safety-critical features for collision avoidance and collaborative actions with humans on industrial or mobile robots.
How to hit a homerun
- Deep reinforcement learning applied to robotic control systems and path planning.
- Hands-on experience processing rich sensor data from LIDAR, RADAR, and cameras.
- Experience with industrial networking protocols OPC UA/EtherCAT/Profibus.
- Industrial PLC programming experience.
Reach out to careers@universeenergy.ai for questions, comments and/or feedback before applying.
