The first DARPA Grand Challenge took place in the Mojave Desert in 2004. It was the first long distance competition for driverless cars in the world. When we first heard of the challenge we immediately thought that KnowledgeScape would be a great platform to provide the intelligence needed to glue together all the sensors, control systems and needed control logic together.

Now, with the introduction of KSX, with it’s peer-to-peer portal communications model it would be easy to define just how KSX could be used to make a car, with all of its systems, truly intelligent and autonomous.

For example KSX could be deployed on multiple Linux single board computers that are in turn connected to appropriate PLC’s to actuate steering, braking, accelerating etc. KSX, through its OPC connectivity, can be used to connect to each of the unique sensors such as radar and a variety of cameras. These local KSX systems can, not only provide intelligent analysis for use at the local level, but could also be used to monitor itself and provide real-time diagnostic analysis at the local level.

Conceptually, each KSX system provides local intelligence to whatever it is connected to. Again, KSX is designed to make anything and everything smart. Crisp and fuzzy rules and even neural network models provide whatever local control is needed. Then additional KSXs are used to interconnect each of the lower level KSXs to provide hierarchical intelligence – or in a sense – providing the glue to put it all together into a system that works towards a common goal.

Breaking systems into small logical pieces and then connecting them together through our portal system creates a fault tolerant system with as much redundancy as desired.

The subject of autonomous vehicles has been much discussed and reported upon the recent years. A few of the autonomous vehicles that are being dreamed about and experimented with include:

• Ore haulage trucks,
• Underwater exploration,
• Drone aircraft,
• Farming vehicles,
• Soil preparation,
• Seed planting,
• Weeding,
• Harvesting,
• Tree pruning,
• Warehouse parts vehicles,
• Etc.

The bottom line for us is that all of these autonomous vehicles need to be coordinated by an intelligent system. KSX is a perfect choice because of the ease of use and how it can be used to group intelligent sub-systems together into higher functioning systems. Partitioning function and control intelligence into small sub-systems is the perfect way to advance the complex engineering and invention necessary to make autonomous vehicles, of all types, an economic reality in the near future.