Future Research Directions¶
This is a brief sketch of some of the future directions that research in evolvable hardware is heading towards.
Mortal Computation¶
This is the idea of using evolvable hardware to create computing systems that are designed to have a limited lifespan, after which they degrade or self-destruct. This concept is inspired by biological systems, where many organisms have a finite lifespan and are programmed to die after fulfilling their reproductive role. There are challenges and opportunities in this area, which has been getting increasing attention recently.
Advanced Concepts: Mortal Computation
Approximate Computing¶
This research direction focuses on using evolvable hardware to create computing systems that can perform computations with a certain degree of error or approximation. This approach can lead to significant reductions in power consumption and hardware complexity, making it suitable for applications where exact results are not always necessary.
Advanced Concepts:Approximate Computing
Autonomous Neural Controllers¶
FPGAs and other reconfigurable hardware can be used to evolve neural network controllers for autonomous systems, such as robots or drones. This approach allows for the creation of adaptive and flexible control systems that can learn and adapt to changing environments.