Kodama Systems has announced a $7 million funding round to speed up the rollout of its supervised autonomous driving technology in US forests. Its equipment is already in commercial use and has hauled more than 100,000 tonnes of timber in six months.
Based in Sonora, California, Kodama Systems develops physical artificial intelligence technology for forestry machines. Its system combines supervised autonomous driving and remote operation. Machines handle repetitive movements automatically, while experienced operators can step in when situations get complex. The company works with several logging operators in the southeastern United States.
Kodama’s machines hauled over 100,000 tonnes of timber in six months
Kodama’s technology is already deployed in commercial logging operations in several southeastern US states. According to the company, machines fitted with its system have hauled more than 100,000 tonnes of timber over the past six months.
Among other customers, Kodama has been working with a client in Georgia for more than a year. These deployments allow the company to gradually adapt its driving functions to different terrains and to the working methods of forestry crews.
The machines involved are mainly skidders, which are used after trees have been felled. Fitted with a large claw called a grapple, they pick up the logs and drag them to an area where they can be loaded onto trucks.
How Kodama’s supervised autonomy and remote operation work
Kodama’s system relies on a hybrid approach. The machine handles the repetitive tasks, while an operator can take over remotely when a situation calls for human intervention.
This setup is aimed in particular at movement over uneven forest ground, where machines have to cross stumps, rocks and ruts. It also lets drivers work without having to stay in the cab at all times.
According to Ben Finfield, head of business development at Kodama, this arrangement addresses the hiring difficulties that logging operators face. Drivers often work long days on remote sites and are exposed to significant physical strain.
The company also points to remote operation as a way to reduce operators’ exposure to accidents involving machine rollovers or falling trees. This is a safety goal put forward by Kodama, not a measured reduction in accidents in the deployments it has announced.
1 terabyte of training data a day from remote operation
Kodama chose to build up automation gradually, starting from remote operation. According to Merit Jenkins, the company’s co-founder and CEO, this strategy was adopted two and a half years ago.
Teleoperation provides data drawn from real situations on logging sites. According to Jenkins, around 1 terabyte of training data is generated every day through the remote operation of the equipment.
This data feeds the development of autonomous driving capabilities across different forest environments. Kodama intends to continue this progression as its technology is installed on more machines.
Integration with two major forestry equipment manufacturers
Rather than building its own forestry machines, Kodama is developing a platform for existing ones. This approach lets operators add remote operation and automation features without necessarily replacing their fleet.
Merit Jenkins says the company has already integrated its technology with two of the world’s three leading equipment manufacturers in the sector.
These partnerships make it possible to use the sensors already fitted to the machines, their fault diagnostics and the manufacturers’ technical expertise.
Kodama now wants to extend its system to other categories of forestry machines, including feller bunchers and loaders. This expansion is part of its development plans, whereas skidders are already a commercial use case.
What the $7 million funding round will pay for
Kodama Systems says it has secured $7 million in new funding, bringing the total capital raised by the company to $14 million.
The round was co-led by Convoy Ventures and Breakwater Ventures, with participation from Blue Forest Asset Management, NYA Ventures, Breakthrough Energy Ventures, Congruent Ventures, Dangerous Ventures and M1C.
The funds are meant to support three priorities: speeding up commercial deployments, strengthening partnerships with machine manufacturers and developing new supervised autonomous driving capabilities for complex off-road terrain.
Possible uses in forest management and wildfire response
Beyond hauling timber, Kodama is considering using its technology for other land management operations. Merit Jenkins cites forest thinning, an activity that can help reduce the amount of flammable vegetation, as well as applications related to firefighting.
The company also holds operational data on its equipment’s real-time location, productivity, fuel consumption and the shocks or vibrations it is subjected to.
According to Jenkins, this information could be used to improve coordination between machines and the organisation of logging sites, beyond automating the movement of individual machines.