Tesla has increased its production capacity for the humanoid robot Optimus tenfold. Although Tesla continues to struggle with Optimus production problems, the company aims to produce 1,000 robots per week by the end of the year.

If the current production momentum is maintained, the company is expected to reach this goal. However, this increase in production volume is not yet sufficient for the robots to operate at full efficiency in real-world conditions.

Production Process Difficulties and Supplier Obstacles

According to information shared by The Information, some disruptions are occurring in the mass production process due to suppliers. Technical problems, particularly in module alignment, can slow down the production line.

The robot’s hands stand out as one of the most complex modules in terms of production and durability. Assembling more than 100 components requires a costly and time-consuming process.

A small assembly defect can cause the modules to need reprocessing. To overcome this problem, Tesla plans to use a replaceable sensor glove instead of replacing all the faulty sensors.

Optimus’s hands must simultaneously provide complex capabilities such as precise movement and tactile sensation. Simulating these features in a robotic system is a very challenging engineering process.

Learning Capacity and Real-World Adaptability

For Optimus to be truly functional in factories, it is not enough for it to simply perform programmed tasks. The robot needs to adapt to unexpected situations when working with objects of different sizes and weights.

For example, to fulfill a worker’s command to pick up boxes and place them on a shelf, it’s not enough for the robot to simply lift the box. It’s also expected to cope with situations such as boxes falling or encountering obstacles on the path.

Today, Optimus can perform tasks for which it has been specifically trained. However, it can exhibit unexpected behaviors when faced with a situation it hasn’t encountered before.

Even some easy tasks can require the robot to be trained for days. To solve this, Tesla is building a large library of basic behaviors.

Basic movements such as grasping, lifting, and walking are used as building blocks for more complex tasks. For this purpose, the company has collected over 500,000 hours of training data.

To expand the dataset, employees are equipped with motion capture suits and cameras are used. This data helps the robot learn new tasks more quickly.

While high production capacity is a valuable step for the commercial future of the project, it is not sufficient on its own. The robots need to function reliably millions of times.

Learning new tasks without long, periodic specialized training is critical to the success of the project. How long do you think it will take Tesla to overcome this complex learning process required for humanoid robots to become widespread in the real world?