BrainChip Podcast: Insights on SpaceTech With European Space Agency ESA.

author avatar

Last updated on 15 May, 2024

BrainChip Podcast: Insights on SpaceTech With European Space Agency ESA.

Listen in as Dr. Tony Lewis, BrainChip's CTO, Laurent Hili (Microelectronics Engineer), and Luis Mansilla (AI Expert) discuss the growing role of AI in space exploration, ESA's investments in this technology, and how BrainChip's Akida chip contributes to this exciting frontier

Have you ever wondered how the next generation of space exploration would be powered? The answer might lie in artificial intelligence (AI), and a new frontier is emerging: AI in space. 

This episode of the Brain Chip podcast explores the vital role of BrainChip in pushing AI to new frontiers through Akida, which was recently launched into space aboard the Transporter 10 spacecraft, alongside Dr. Tony Lewis, BrainChip’s CTO, and experts from the European Space Agency (ESA) Laurent Hili and Luis Mansilla.

The Rise of New Space 

The podcast begins by highlighting the dramatic shift in space exploration. Gone are the days of solely government-driven endeavors. Private companies like SpaceX are making space more accessible, leading to a new era of "New Space." This increased activity necessitates advancements in spacecraft autonomy and mission efficiency.

One solution gaining traction is AI. Implementing AI systems directly on spacecraft can automate complex tasks, freeing human astronauts for critical decision-making and exploration activities. Additionally, onboard AI can analyze data in real time, enabling faster responses to unforeseen situations and optimizing mission parameters. However, the podcast emphasizes the challenges of using traditional AI hardware in space.

Neuromorphic Computing: A Promising Path Forward

The podcast also explores the complexities of implementing AI in space. One of the most significant hurdles is power consumption. Traditional AI hardware can be power-hungry, a major constraint for spacecraft with limited energy resources. As discussed, traditional computing struggles with the power constraints of space. 

This is where neuromorphic computing emerges as a game-changer. Unlike conventional processors that rely on the von Neumann architecture, neuromorphic computing takes inspiration from the human brain.

Here's how neuromorphic computing breaks the mold:

  • Mimic the brain's structure for parallel processing and potential efficiency gains.

  • Consume significantly less power, which is crucial for space where every watt counts.

  • Can be tailored for specific AI tasks, further boosting efficiency.

While still young, neuromorphic computing holds immense promise for space AI.

This is where AI comes in. Artificial intelligence has the potential to revolutionize space exploration by enabling:

  • Onboard decision-making: AI algorithms could analyze sensor data and make real-time decisions without relying on constant communication with Earth, which vast distances and time delays can hinder.

  • Autonomous operations: Spacecraft equipped with AI could perform tasks like navigation, anomaly detection, and even scientific data analysis with minimal human intervention.

  • Data reduction and filtering: AI can sift through vast amounts of data collected by space instruments, identifying valuable information and reducing the amount that needs to be transmitted back to Earth, saving precious bandwidth.

ESA's Multifaceted Approach to AI in Space

The podcast goes beyond just neuromorphic computing, showcasing the multifaceted approach of the European Space Agency (ESA) to integrating AI into various space missions. As highlighted by ESA, they are exploring AI applications across different departments, including:

  • Beamforming and anti-jamming technologies for improved communication and signal strength.

  • Maritime applications such as ice bear detection for environmental monitoring.

  • Earth observation tasks like analyzing data from the new SC of the Cloud.

They also discussed ESA's adoption of the NASA Technology Readiness Level (TRL) system to assess the maturity of AI technologies for space use. This standardized approach helps prioritize development efforts and ensures that technologies are adequately tested before flight.

BrainChip's AK100

The podcast discusses BrainChip's AK100 chip and its potential role in ESA's space missions. BrainChip reveals that the AK100 chip is currently at TRL 3, indicating it's in the early stages of development. Further work is required to harden the technology for space environments, including rigorous testing for radiation resistance and mitigation strategies.

Conclusion: A Collaborative Future for AI in Space

The BrainChip podcast episode offered a glimpse into the exciting and complex world of AI in space. While challenges remain, particularly regarding power consumption and technology validation, the potential benefits of AI for space exploration are undeniable. The collaboration between ESA and BrainChip exemplifies the innovative spirit driving this new frontier.